advanced inspection project final report_tcm174-283988

ADVANCED INSPECTIONS PROJECT INNOVATIVE TECHNOLOGIES TO IMPROVE HIGHWAY CONDITION MONITORING

RWS-DVS

28 September 2009

D03031/CG9/0M0/001999/ep

D03031.001999

ADVANCED INSPECTIONS PROJECT

D03031/CG9/0M0/001999/ep ARCADIS 2

Contents

Summary ______________________________________________________________________ 4

1 Introduction _________________________________________________________________ 7

1.1 Project Background _______________________________________________________ 7

1.2 Project Focus ____________________________________________________________ 8

1.3 How should this report be read? ____________________________________________ 9

2 Research approach _________________________________________________________ 10

3 Pavements ________________________________________________________________ 12

3.1 Introduction ___________________________________________________________ 12

3.2 Inspection Requirements _________________________________________________ 12

3.3 Current Practice ________________________________________________________ 13

3.4 Leading Practice ________________________________________________________ 15

3.5 Opportunity I: Detection of Pavement Defects _______________________________ 20

3.5.1 ‘Car as Sensor’ ___________________________________________________ 21

3.5.2 Mobile 3D Video Mapping _________________________________________ 24

3.5.3 Unmanned Aerial Vehicles _________________________________________ 26

3.5.4 Spaceborne Techniques ___________________________________________ 27

3.5.5 Ground Penetrating Radar _________________________________________ 28

3.6 Evaluation of Opportunity I Techniques _____________________________________ 29

3.7 Opportunity II: Predicting Pavement Condition to Prevent Defects _______________ 30

3.7.1 ‘Smart Dust’ _____________________________________________________ 31

3.7.2 Fingerprinting Pavement Composition _______________________________ 33

3.7.3 Hyperspectral Techniques __________________________________________ 34

3.7.4 ‘A recipe for good asphalt’, ASPARI _________________________________ 36

3.8 Evaluation of Opportunity II Techniques ____________________________________ 37

4 Structures _________________________________________________________________ 39

4.1 Introduction ___________________________________________________________ 39

4.2 Inspection Requirements _________________________________________________ 40



4.5 Opportunity III: Supporting In-Field Inspection Technologies ____________________ 45

4.5.1 Tagging Structure Elements ________________________________________ 45

4.5.2 Cable Integrity Using Electromagnetism ______________________________ 48

4.5.3 Digital Photography_______________________________________________ 49

4.6 Evaluation of Opportunity III Techniques ____________________________________ 50

4.7 Opportunity IV: ‘24/7’ Monitoring of Defects ________________________________ 50

4.7.1 In-Situ Sensors ___________________________________________________ 51

4.8 Evaluation of Opportunity IV Techniques ____________________________________ 52

4.9 Opportunity II: Prevention of Defects _______________________________________ 53

4.9.1 Smart Paint – C-Cube _____________________________________________ 54


D03031/CG9/0M0/001999/ep ARCADIS 3

4.9.2 Fibre Optics _____________________________________________________ 54

4.9.3 Deformation Sensors ______________________________________________ 56

4.9.4 InSAR Technology ________________________________________________ 56

4.10 Evaluation of Opportunity V Technologies __________________________________ 59

5 Objects ___________________________________________________________________ 61

5.1 Introduction ___________________________________________________________ 61

5.2 Inspection Requirement __________________________________________________ 61



5.5 Opportunity VI: Automated Inspections_____________________________________ 63

5.5.1 Mobile 3D Laser Mapping _________________________________________ 63

5.5.2 Mobile Video Mapping of Objects ___________________________________ 66

5.6 Evaluation of Opportunity VI Techniques ____________________________________ 68

6 Embedding new technologies _______________________________________________ 70

6.1 The Current Situation ___________________________________________________ 70

6.2 The Role of the Private Market Sector ______________________________________ 72

6.3 Management Changes __________________________________________________ 75

7 Conclusions _______________________________________________________________ 77

8 Recommendations _________________________________________________________ 80

8.1 Introduction ___________________________________________________________ 80

8.2 Most Promising Inspection Technologies ____________________________________ 80

8.2.1 ‘Car as Sensor’ ___________________________________________________ 80

8.2.2 ‘Smart Dust’ _____________________________________________________ 81

8.2.3 Tagging Assets In Relation To 3D Inspection Technologies _______________ 81

8.2.4 Mobile 3D Video and 3D Laser Mapping _____________________________ 82

8.2.5 Radiometric Fingerprinting _________________________________________ 83

8.2.6 InSAR Technology ________________________________________________ 83

8.3 Embedding New Technologies at RWS _____________________________________ 84

Annex 1 List of People Interviewed ______________________________________________ 87

Annex 2 Aran _______________________________________________________________ 89

Annex 3 ‘Car as Sensor’ _______________________________________________________ 90

Annex 4 Scoring of Techniques _________________________________________________ 92

Annex 5 Evaluations __________________________________________________________ 95

Annex 6 Review & evaluation meeting __________________________________________ 113


D03031/CG9/0M0/001999/ep ARCADIS 4

Executive Summary

This report provides the results of the Advanced Inspections Project (AIP) – which is part of

the ‘Innovation Projects Highway Maintenance’ (Dutch: ‘Innovatieprojecten Wegonderhoud’,

IPW) programme of RWS-DVS. The project focus was on innovative inspection methods and

investigated a number of opportunities that could deliver improvements in (a) better asset

knowledge (b) safety, (c) efficiency and (d) reduced hindrance (vehicle delay). Technologies

and information needs were discriminated for Pavements (Chapter 3), Structures (Chapter

4) and Objects (Chapter 5), respectively. Next to the identification of innovative technologies

also attention is being paid about how to embed innovative inspection technologies within

the RWS organization (Chapter 6). The Advanced Inspections Workshop (Reid and

Oostrom, 2009) held at the 8th of July and organised by RWS-DVS reveals interesting

information needs that have been used to provide some recommendations about the most

promising technologies (Chapter 7).

The investigation reveals that - except for some spaceborne and airborne sensor technologies

that are under investigation - the most promising technologies are still terrestrial based. This

is mainly due to the low resolution of the spaceborne imageries and the low flexibility of

both sensor types. Although these technologies do not cause traffic hindrance, the problem

is that observations have to be done during daylight meaning that pavement is often

obscured due to the high traffic intensity. The airborne and spaceborne technologies might

be of interest though for hot spot detection over large (nationwide) areas and for the

inventory of road assets.

With the terrestrial based sensors there is a clear tendency in the sensor development to

retrieve digital information at traffic speed for further elaboration of the information in the

office. This has advantages compared to the current visual inspections. Observations will

become more objective, are safer for the inspectors and will lead to less traffic hindrance. If

properly archived, the digital information can be used for retrospective analysis of defects or

wearing processes as well. The automatic extraction of features by means of optical images

or similar sensing sources is still under development. This means that visual inspection of

the images is still necessary. Another clear tendency is the retrieval of 3D information. 3D

techniques like Mobile 3D Laser and Video Mapping are better suitable for representing

properties of an object then 2D techniques. 3D feature manipulation is being developed by

CAD & GIS software systems, it looks that 3D design & build information can be connected

or transferred to spatially based 3D asset management systems. By tagging the road assets

in the field (for instance structure components) and associate them with properties

(‘attributes’) the components can be traced in the field and new findings about the condition

of an object can be (real time) updated in the databases.

Current, and the majority of innovative, inspection technologies still focus on the detection

of physical defects, rather than they support in the prevention of physical wearing by

monitoring the deterioration processes in time. However, there are some promising

technologies like ‘smart paint’ and pavement ‘fingerprinting’ technology that measure

changes in the composition of the material and can tell about the status of it. However, for

the Pavement ‘fingerprinting’ technology more knowledge should be obtained to


D03031/CG9/0M0/001999/ep ARCADIS 5

understand the relation between composition changes and the physical deterioration

process. Key question for the asset manager is how to avoid physical defects and replace the

material at the right time. That will considerably decrease cost-benefit ratios and contribute

to sustainable road management.

A very interesting new development is the ‘Car as Sensor’ concept. ‘Car as Sensor’ implies a

form of participatory mapping that deploys the public car fleet to assess road conditions.

Presently ‘Car as Sensor’ technologies still particularly focuses on traffic flow management

and in-car safety. However, some R&D projects now also focus on how car sensors such as

active suspension sensors, brake sensors, and other car state sensors can be of help to

measure road conditions. When this information becomes available to road managers, it

basically means that millions of sensors are available round the clock, seven days a week

(‘24/7’) measuring road conditions network wide. This may sound still like a distant future,

however, with a stepwise introduction of the technique it will already be possible to start

testing the potential of the technology (see also Section 3.5.1 and Annex 3: Review &

Evaluation Meeting Notes, Delft, 16 September 2009). This ‘in-house’ pilot would for

instance overcome the sensitive issue of privacy.

In Table 1 below, seven promising techniques are listed that were identified based on a

scoring that looked at data acquisition speed, technical performance, traffic management

needs and cost. The latter is merely a rough estimate for the R&D technology, covering the

equipment cost, but excluding the cost of (technical) personnel needed and the post-

processing of data. The opportunity column describes how the technique will benefit

inspections, which can either through detecting of defects, preventing defects, round the

clock (‘24/7’) monitoring or through supporting inspections. These opportunities are

described in more detailed in Chapters 3, 4 and 5 below. Summary evaluations of all

innovative techniques can also be found in Annex 5.

Table 1. Most promising techniques based on a scoring per asset category and opportunity.

Pavements

Technology Opportunity

‘Car as Sensor’ Detection

Mobile 3D Mapping Detection

‘Smart Dust’ Prevention

Structures


In-situ sensors ‘24/7’ monitoring

RFID tagging Support inspections

Objects


Mobile 3D Mapping Detection

The above described technologies have one thing in common, they will create large amounts

of data that need to be processed and analysed. New technologies like ‘Car as Sensor’ will

not only provide a huge amount of data but it also need to be handled in a very short time

span if road information condition information services will become part of the information

chain. From a technical point of view this means that databases should be able to handle


D03031/CG9/0M0/001999/ep ARCADIS 6

such an amount of data, data filtering procedures need to be developed and wireless

communication networks should be able to handle the data flow. From an organizational

point of view it means that RWS need to engage with the public and it should strive to a

closer collaboration with the (commercial) information and communication society as well

as with the automotive industry which is at the front end of developing new in car sensor

technologies. However, as RWS is currently rethinking its position with respect to the

private market sector regarding road asset management and inspections (Partner

programme Infrastructure Management; PIM), it might be the right time to consider how

innovative technologies best can be embedded within the organization and what will be the

impact at a technical level but also on an organizational level.

The RWS policy ‘The Market unless…’ seems to indicate an increased involvement of the

market regarding road asset monitoring and maintenance. Therefore, it is currently

uncertain what position RWS will take regarding the stimulation of innovation. During the

concluding ‘Review & Evaluation Meeting (See also Annex 6) it was further concluded that

for some of the most promising technologies, business cases need to be worked out with

detailed and complete cost-benefit assessments.


D03031/CG9/0M0/001999/ep ARCADIS 7

CHAPTER

1 Introduction

1.1 PROJECT BACKGROUND

Good traffic flow in the Netherlands is essential for a healthy economy. The national

motorway network manager, Rijkswaterstaat, has the important task of keeping the

Netherlands ‘on the move’. The aim is to make sure that journeys are safe and reliable for

goods and people. In order to achieve this, the Netherlands’ main road network needs to be

kept in good condition. As traffic intensity increases this puts more pressure on

maintenance needs and shortens the periods in which this maintenance can be carried out.

The pressures on the network and the objective for good traffic flow prompted the Dienst

Verkeer en Scheepvaart (DVS) in the Rijkswaterstaat to launch a programme of Road

Management Innovation Projects (Innovatieprojecten Wegonderhoud, IPW). The projects

within the IPW are intended to reduce the hindrance to road users through reducing the

need for road maintenance and by increasing the efficiency of maintenance. Therefore, the

aim is to cause less hindrance despite more traffic.

The main objectives of the Advanced Inspections Project - which is part of the IPW - is to

focus on innovative inspection methods that could deliver improvements in (a) better asset

knowledge (b) safety, (c) efficiency and (d) reduced hindrance (vehicle delay). The project

has sought opportunities to share knowledge with other road authorities, research

institutions and wider industry. New opportunities and ideas have been evaluated to

develop recommendations for practical application or testing of improved techniques. The

project is jointly managed and resourced by the Dutch Rijkswaterstaat and English

Highways Agency, who are working with technical consultants from ARCADIS in the

Netherlands.

The Advanced Inspections Project as described in this report is the follow up of a

preliminary effort that first of all looked at the project setup and then looked at impact of

inspections on traffic flow. The results of these first two phases were used to set the scope

for the next three phases that are described in this report (also see Figure 1). The conclusion

of the first two phases was that inspections in fact cause relatively little hindrance or vehicle

delay. More important therefore was it to focus on issues like improved safety for

inspectors, more efficient inspection techniques, and also think about collecting enhanced

data and ways to use the data in a better way. Through better knowledge about the current

condition of the asset, road asset managers will be better able to:

• Predict - when a functional failure of an asset is going to happen

• Plan - when maintenance/renewals are needed. This assist with financial planning and

planning network access.


D03031/CG9/0M0/001999/ep ARCADIS 8

• Prioritize - which maintenance work is urgent, what can be postponed and what needs

to be monitored more closely.

• Patch - make lifetime extending repairs to an asset.

Fig.: 1. Phase 1 & 2 were the pre-phases of the AIP project which encompasses Phase 3

through 5.

1.2 PROJECT FOCUS

Also through Phase 2 interviews with stakeholders, opportunities were identified that could

potentially improve the inspections and maintenance processes. These opportunities can be

grouped into two categories ‘Automated Inspections’ and ‘Intelligence based asset

management’. These opportunities are shown below:

Intelligence-led Asset Management

Have the ability to continuously monitor the asset, rather than undertaking inspections

periodically at set intervals

Base maintenance and renewal decisions on condition measurements rather than

assumptions about the state of the asset

Understand the condition of the structure below the surface (concrete bridges and road

under the pavement)

Monitor trends over time in the condition of the asset to assist planning and decision-

making.

Inform a risk-based approach, focusing on areas that are high-risk and require greatest

attention

Automated Inspections

Reduce exposure of inspectors to dangerous working conditions

Reduce the reliance on human observations at traffic speed

Use objective measures, rather than subjective judgments

Automate processes to improve efficiency - fewer manual processes that are often time-

consuming and costly

The scope of the AI project coincides with the 2025 vision of FEHRL (Forum of European

National Highway Research Laboratories):


D03031/CG9/0M0/001999/ep ARCADIS 9

‚Monitoring of road condition will be achieved by the use of condition monitoring vehicle travelling

at traffic speed, and by implanted sensors in road pavements and structures to register and record

condition automatically. Sensors will be used to give advance warning of structural deterioration and

enable inspection intervals to be increased thus reducing the associated traffic disruption‛.

(FEHRL: ‘Vision: Road Transport in Europe 2025’)

1.3 HOW SHOULD THIS REPORT BE READ?

The report is meant for RWS asset managers and people responsible for inspections and

maintenance of pavements, structures and other road objects. The scope of this report is

therefore to address the observed problems and requirements and relate them to possible

technical innovations that can contribute to (a) better asset knowledge (b) safety, (c)

efficiency and (d) reduced hindrance (vehicle delay). The main sections therefore start with

the road assets that need to be inspected: pavement, structures and road objects like barriers

and gantries1. In the respective sections current practice, leading practice and opportunities

for innovation will be discussed (see also Figure 2). As current practice technologies are

already in use by RWS it is assumed they are known to the stakeholders. Therefore, the

report will only briefly refer to these technologies.

Fig 2. Scope of the investigation.

Chapter 6 provides information about embedding new technologies within the RWS

organization. An important subject as RWS is restructuring its organization and its relation

with the private market sector regarding road maintenance and inspections. The last section

will provide a set of Recommendations on the subject of the most promising technologies.

These recommendations are based on the discussions with ARCADIS experts and will also

be discussed with RWS experts before being agreed.

1 DVM systems are not considered in this report as DVM performance inspections are done electronically and through software applications (eg RWS Da Vinci Project) and fall outside the scope of this project.


D03031/CG9/0M0/001999/ep ARCADIS 10

CHAPTER

2 Research approach

Commissioned by the Centre for Transport and Navigation (RWS-DVS), part of the Dutch

Ministry of Transport, Public Works and Water Management (Rijkswaterstaat), ARCADIS

has played an advisory role in the Advanced Inspection Project (AIP). AIP resides under the

‘Road Management Innovation Programme’. The project involved investigating possible

new inspection methods that do not inconvenience traffic and are safer for inspectors. In

addition, the improvement of asset knowledge and efficiency in gathering information were

prerequisites when looking for new inspection techniques.

It has been agreed that ARCADIS will look into new inspection methods for pavements,

structures (e.g. bridges) and road objects (e.g. safety barriers and road markings). Each asset

category is treated in a separate chapter. The following structure has been adopted

throughout this report; first, inspection requirements, current practice, and leading practice

are described, followed by opportunities for innovation. In agreement with RWS-DVS,

ARCADIS decided not to go into much detail for each technique as the scope of the

investigation is relatively wide.

As part of the investigation ARCADIS held several interviews with employees of

Rijkswaterstaat, international highway maintenance organizations, scientists and engineers,

contractors and manufacturers of sensors and inspection systems (see Annex 1: List of

interviewed people and affiliations). In addition, ARCADIS collected literature and reports

that describe innovative techniques that could possibly be used in future inspections. A

further source of information was the Advanced Inspections Workshop (Reid & Oostrom,

2009) which took place in Utrecht on 8th July 2009 and was organized by RWS. It involved a

variety of people from across RWS as well as representatives from the English Highways

Agency, research institutes (such as TNO, TRL and Deltares) and other organisations with

involvement in asset management (such as ProRail). During the workshop RWS-DVS and

ARCADIS were able to collect additional information on requirements and opportunities by

having RWS staff and ‘the market’ or ‘resource’ persons discuss the ‘ideal future’ of

inspections and their ‘information needs’. In an Expert Meeting and a separate Review &

Evaluation Meeting the preliminary conclusions and recommendations were further

adjusted and the feasibility of innovations for future inspections was discussed.

The project workflow is also illustrated in Figure 3, starting with identifying and

interviewing the key players. Important points of communication between ARCADIS and

RWS-DVS were the (bi-weekly) meetings. At the end of Phase 3&4, a first draft report was

reviewed in an ARCADIS expert meeting. The results of the meeting discussions lead to a

full draft report that was presented in a Review & Evaluation Meeting where RWS experts

on pavements and structures were present, resulting in a revised final report.

http://en.wikipedia.org/wiki/Ministry_of_Transport,_Public_Works_and_Water_Management_(Netherlands)


D03031/CG9/0M0/001999/ep ARCADIS 11

Fig.: 3. Diagram illustrating the workflow, points of communication between stakeholders and

important milestones.


D03031/CG9/0M0/001999/ep ARCADIS 12

CHAPTER

3 Pavements

3.1 INTRODUCTION

This chapter covers pavement inspection requirements, current practice, leading practice

and opportunities for innovation in pavement inspections. The opportunity sections are

concluded with an evaluation of possible innovations.

3.2 INSPECTION REQUIREMENTS

Better understanding of the pavements conditions and timely observation of deterioration

are necessary to optimize the Dutch road pavement network. Whereas currently RWS is

principally observing the physical defects of asphalt deterioration (see Table 2), optimally

RWS would like to prevent such stage of road conditions by a timely replacement of asphalt.

It is for this reason that this chapter identifies possible technologies that can be helpful to

detect deterioration effects and technologies that might provide a better understanding of

the pavement condition before the pavement physically deteriorates.

Table 2: Pavement inspection requirements and currently used techniques (source: RWS).

Inspection Reason

Surface friction/skid resistance Skidding/breaking distance

Rutting Trucks

Transversal and longitudinal profiles Construction

Skid resistance, wet HA Skidding/breaking distance

Road surface noise External vehicle noise

Construction layers Conditions underlying layers

Deflection response Load bearing capacity, condition under surface

Cracks (depth) Sub surface

Cracks (surface) Holes, uneven road

Raveling ZOAB (porous asphalt). Failure of bond between aggregate and binder in combination with ageing

Dirt Drainage ZOAB

'Local defects' Pot holes, subsidence, edge deterioration, fretting, loss of chippings, fatting up

Vegetation Overgrown, sight-lines


D03031/CG9/0M0/001999/ep ARCADIS 13

Pavement condition inspections are based in The Netherlands on well-defined and

standardized criteria. Visual inspections methods are developed by CROW (Centre for

Regulation and Research in Ground, Water and Road Construction and Traffic technique)

and are described in detail in CROW Manual 146a. Global visual inspections are described

in CROW Manual 146b.

3.3 CURRENT PRACTICE

Currently three types of pavement inspections are done at traffic flow speed: the measuring

of rutting depth, of transverse and longitudinal evenness (International Roughness Index,

IRI), and of skid resistance. Ravelling, cracking and detection of other pavement defects are

done by visual inspections. The same holds for spotting foreign objects and checking

vegetation height. These and other (local) pavement aspects including dirt, are currently

inspected visually by means of 100 m (hectometre) transects (Table 3). In green those

techniques are mentioned that are done at traffic flow speed. In yellow are marked

techniques that are done static or at low speed and by visual inspection. For example, to

determine the structural integrity and level of ageing of pavement layers, sample cores are

drilled. Coring and deflection response tests are currently done static or at very low speed,

which causes traffic hindrance and endanger the safety of inspectors.

Table 3. Inspections and currently used techniques (source: RWS)

Inspect Currently used technique

Skid resistance Wheel towed behind car on a wet surface (ROAR)

Rutting ARAN-3 Profiling laser

Transverse and longitudinal profiles International Roughness Index

ARAN-3 Profiling laser

Cracks (surface) Semi-automated Video + detection software

Deflection response FWD Falling Weight Deflectometer

Construction layers Coring, samples

Cracks (depth) Asphalt cylinders

Ravelling Visual

Dirt and grease Visual

Local defects Visual

Vegetation Visual

Visual Inspections

At this moment, in the Netherlands most pavement defects are checked visually by sub-

contractors that drive on the hard shoulder with low speed (~20 km/h). They record

problems of greasing, cracking and especially ravelling, providing RWS with information at

hectometre level. However, with visual inspections objectivity is a problem; different people

will come back with different scores for the same stretch of road. In addition, some 20% of

highways cannot longer be inspected from the hard shoulder due to new safety regulations.

http://www.crow.nl/


D03031/CG9/0M0/001999/ep ARCADIS 14

Automatic Road ANalyzer (ARAN)

In the Netherlands RWS-DID currently uses the 3rd generation ARAN (RoadWare

Corporation, USA) vehicle was put into operation on 15 May 2007 (see Figure 4; also see

Annex 2: ARAN for further details). ARAN measures: rutting depths, International

Roughness Index (IRI) and transverse profiles. The ARAN uses two cameras synchronized

with a strobe illumination system, with each camera covering about half-width of a

pavement lane. Since late 1996, WiseCrax software is available for the semi-automated

survey of pavement surfaces, focusing on cracking of asphalt. Estonia uses DynaCrack, the

Austrian Institute of Technology (AIT) uses ‘RoadSTAR’ featuring automated crack

detection ≥ 1mm and the UK has a HARRIS vehicle. In all cases, the detection of cracks in

pavements is supported by specialized software, but humans are still needed to confirm and

map the pavement problems found by the computer.

Use of ARAN video frames was tested for inspections; however, when using high speed

video, the images proved not clear enough. The latter leads to poor results in scoring the

level of ravelling when compared to visual inspections done from a car driving on the hard

shoulder. During the Advanced Inspections Workshop (Reid & Oostrom, 2009), RWS

inspectors added that the lack of scale in video imagery makes a proper assessment of

pavement conditions more difficult than when seen from a car directly.

Fig.: 4. The 3rd generation ARAN vehicle (source: RWS-DID).

Road Analyzer and Recorder (ROAR)

In the Netherlands the wet friction (Dutch: ‘natte stroefheid’) of roads is measured according

to a method that was developed in the 50’s by then National Road Construction Laboratory

(Dutch: ‘Rijkswegenbouwlaboratorium’). The method, described in RAW test 150, pulls a wheel

at a measurement speed of 50 kph, which is slowed down 86% when compared to a free

rolling wheel. The wheel is outfitted with a PIARC-P measurement tyre, without profile

(worst case scenario). Under this tyre, a volume of water is sprayed that would form a water

layer of 0,5 mm thickness on a surface without texture. It is obvious that the water depth in

practice depends on the macro texture and porosity of the road surface. This is similar to

conditions during a rain shower. The aforementioned texture is a measure for the speed

dependence of the skid resistance and is measured with texture lasers or the sand patch

method (RAW test 111). On the main road network skid resistance is monitored with the

ROAR vehicle (Figure 5) of RWS-DWW. This system can measure the entire skid curve from

http://roadware.com/

http://roadware.com/


D03031/CG9/0M0/001999/ep ARCADIS 15

free rolling to a blocked wheel or if needed with a fixed value for skid resistance.

Measurement speed can be adapted to traffic conditions and a water layer of 0,5 mm is used.

Source: KOAC-NPC).

Fig.: 5. RWS-DWW ROAR test vehicle, note the water tank and the PIARC tyre on the back (source:

RWS-DID).

It was found that internationally, and also among European countries different methods are

used to measure skid resistance of pavements. Currently a programme is underway to

harmonize all measurements to comply with European legislation that could be effective in

2012.

3.4 LEADING PRACTICE

In Australia automated detection of cracks is done by processing line scanning video by

RoadCrack software. The Common Wealth Scientific and Industrial Research Organisation

(CSIRO) was the first to design an automated road crack detection system that can identify

cracks (>1mm width) while operating at highway speed (>80 km/hr). It was the outcome of

a project for the Roads and Traffic Authority of New South Wales. RoadCrack, is a fully-

automated system. It combines advances in:

Machine vision

Parallel computing

Artificial intelligence

Image analysis.

High-speed cameras are mounted under a vehicle to collect digital images of the pavement

surface while the vehicle is moving at highway speed. A special reflector system focuses

light to illuminate tiny cracks. High-resolution images are collected for small sections of

pavement and then consolidated into bigger images that cover half-metre intervals of the

road surface. CSIRO also developed algorithms and computer hardware for extracting the

relevant information in real time (see also Figures 6a and b).

http://www.koac-npc.com/Flex/Site/Download.aspx?ID=1330


D03031/CG9/0M0/001999/ep ARCADIS 16

Fig.: 6a. Digital image of road pavement. Fig.: 6b. Detected crack patterns.

During operation, images are passed to a processing module that uses sophisticated image

analysis algorithms to classify cracks as they are detected. Cracks are reported according to

type, severity and extent.

RoadCrack detects cracks as fine as one millimetre in width, while travelling up to

105 kilometres per hour. It can also operate at night as the system provides its own light

source. The objective data from RoadCrack provides valuable input into management of

road pavement assets, saving the NSW State government roads authority tens of millions of

dollars each year in road maintenance costs (CSIRO).

For ravelling and pavement defects, other than cracking, no automated detection techniques

have been found in literature. Researchers at the Leuven University have good hope that

they can develop an algorithm that picks up ravelling from frame camera imagery. It

depends on the detail that is required if Mobile 3D Video Mapping (section 3.5.2) or any

other technique can detect ravelling successfully.

In the US the Transport Research Bureau (TRB) presented the 2nd

Strategic Highway

Research Programme (SHRP2) results in an international conference2. A great number of

pavement inspection techniques were identified as leading practice in the US (Source: US

TRB, SHRP2-TTI3). The results of the investigation are summarized below.

Real Time Automated Distress Data Collection

Real Time Automated Distress Data Collection is the collective name of the collection of

pavement distress data with automated processes for use in pavement management and

design activities, especially for planning and designing maintenance and rehabilitation

treatments. However, it has been difficult to determine the quality of this data in terms of

accuracy, precision, and resolution. Problems contributing to this difficulty include the lack

of mutually accepted standard distress definitions. TRB is doing ongoing research to reduce

distress data collection into its most fundamental form that could then be interpreted

differently by each agency. Distress data collection systems being used in the USA can be

compared to the ARAN (Netherlands) or HARRIS (UK) vehicles. Manual interpretation

systems widely used but they are expensive. Automated systems (some real time) under

2 2nd International Symposium on Non-Destructive Testing for Design Evaluation and Construction

Inspection, 25 April 2008 Ljubljana, Slovenia

http://www.csiro.au/solutions/psaa.html

http://www.trb.org/AboutTRB/Public/SHRP2.aspx

http://onlinepubs.trb.org/onlinepubs/shrp2/TTI3.pdf



D03031/CG9/0M0/001999/ep ARCADIS 17

development in various DOT’s3 and in private companies. The application is foremost to

make cracking estimates on flexible pavements.

Inertial profilers

Technology principles: inertial profilers are based on the use of multiple laser arrays to

measure the transversal profiles of a lane. Most Departments of Transportation (state run

organisations) use inertial profilers to make high-speed inventories of pavement smoothness

and for acceptance testing of as-built pavement smoothness. Majority of specifications are

for hot mix asphalt (HMA) testing only. Used on PCC pavement construction in a few states.

Initial interest exists on acceptance testing of base smoothness. Advantages are: provide

objective measures of ride quality, profile data useful for various applications, high-speed

and 100% coverage of test wheel path, mature technology - used in-house by most DOTs. A

limitation is that it provides the user with relative profile measurements only. Barriers to

further implementation: Cost could be a factor, lack of training on proper operation and use,

lack of understanding on what equipment can or cannot do.

MIT-Scan

Technology principles: MIT -SCAN technology operates using the principles of the Electro-

Magnetic Tomography (EMT). EMT is the technique enabling to visualize spatial

distribution of electrical impedance (or conductivity) inside the object. The device uses

voltage measurements on the object's surface when the electric current passes through the

volume, as initial data for the image reconstruction. By means of a sensor field, a high

sampling rate and an accurate distance measurement in the moving direction of the

measuring device, magnetic answering fields are recorded. EMT technique locates position

and depth of placement of dowel bars4. Advantages are, it is easy to use (Washington DOT

and FHWA experience), can be used to monitor dowel bars during construction, it is not

affected by presence of water or change in moisture content. Limitations: presence of near

metal objects can introduce errors, upper limit of 190 mm for depth, evaluation limited to

bar types included in parameter files. Barriers to implementation: lack of specifications on

use for construction, lack of understanding on what equipment can or cannot do.

Acoustic Emissions

Technology principles: Acoustic Emissions is a method and apparatus for analyzing

selected properties of a ferromagnetic material by magnetically inducing acoustic emissions

in the material. The measured acoustic emission responses are compared to each other and

to standard responses. Comparing the ratio between acoustic emission responses with

standard response ratios for the same material characterizes the condition of the material.

Not known if already applied by highway agencies or DOT’s. Applications: detecting the

debonding of the deck from the girders, detection of delaminations, monitoring the tip

velocity of a propagating crack. Advantages are, it is a passive technique, no excitation of

ultrasonic signals is required. Limitations, history of acoustic emission signals cannot be

caught on existing structures. Estimation of the structural integrity of existing structures is

difficult.

3 Department of Transportation

4 A dowel bar retrofit is a method of reinforcing cracks in highway pavement by inserting steel dowel

bars in slots cut across the cracks


D03031/CG9/0M0/001999/ep ARCADIS 18

Air Coupled Ultrasonic Testing

Technology principles: When sound passes across an interface between two materials only a

proportion of the sound is transmitted, the rest of the sound is reflected. The proportion of

the sound that is transmitted depends on how close the acoustic impedance of the two

materials matches. Air coupled ultrasonic technology is measuring these characteristic of

different materials.

Applications are: identification of thickness of pavement, identification of complex modulus

of elasticity of pavements, non-destructive testing of bridge decks, non-destructive testing of

tunnels. Advantages are, it is a non-contact technique, it has the potential for high speed

measurement and evaluation. Limitations are: high impedance mismatch using air as a

coupling medium, the generated and detected waves have a low signal; amplitude is

relative to contact transducers. It is not known if already applied by highway agencies or

DOT’s.

Laser Scanning

Agencies considering or using this technology: California, Minnesota, Texas, Washington,

FHWA and various consultants. Applications are monitoring movements or deformations,

mapping of highway assets, tunnel inspections and scour measurements. Advantages of the

technique are that it captures ‘as-is’ conditions, provides 100% coverage, provides

continuous monitoring for change detection, and allows measurements of hazardous or

inaccessible environments. Limitations are that targets must have clear line of sight, and

atmospheric effects (weather). Barriers to implementation are the high cost that could be a

factor, standards needed for comparing systems and automated processing of scan data is

still limited.

Magnetic Particle Testing

Technology principles: The Magnetic Particle Inspection method is a non-destructive

method for locating surface and sub-surface discontinuities in ferromagnetic material. It

depends for its operation on the fact that when the material or part under test is magnetized,

discontinuities that lie in a direction generally transverse to the direction of the magnetic

field, will cause a leakage field, and therefore, the presence of the discontinuity, is detected

by use of finely divided ferromagnetic particles applied over the surface, some of these

particles being gathered and held by the leakage field, this magnetically held collection of

particles forms an outline of the discontinuity and indicates its location, size, shape and

extent. Magnetic Particle Testing does not indicate the depth of the imperfection. A

magnetic flux is send through the material. It is considered as a quick and reliable technique

for detection and location of example surface cracks. Applications are identification of

corrosion rate and extent in rigid pavements, bridge components, and tunnels. Advantages

of Giant Magnetoresistive (GMR) sensors, they offer a promise for magnetic sensing of rate

of corrosion and extent and concrete without having direct contact with the steel. The

sensors have high sensitivity, are inexpensive and small. No limitations are currently known

yet.

Ground Penetrating Radar

Technology Principles: Ground-penetrating radar (GPR) is a geophysical method that uses

radar pulses to image the subsurface. This non-destructive method uses electromagnetic

radiation in the microwave band (UHF/VHF frequencies) of the radio spectrum, and detects

the reflected signals from subsurface structures. A dielectric contrast must exist between

http://en.wikipedia.org/wiki/Geophysics

http://en.wikipedia.org/wiki/Radar

http://en.wikipedia.org/wiki/Electromagnetic_radiation

http://en.wikipedia.org/wiki/Electromagnetic_radiation

http://en.wikipedia.org/wiki/Microwave

http://en.wikipedia.org/wiki/Band_(radio)

http://en.wikipedia.org/wiki/Ultra_high_frequency

http://en.wikipedia.org/wiki/VHF

http://en.wikipedia.org/wiki/Radio_spectrum


D03031/CG9/0M0/001999/ep ARCADIS 19

layers, therefore new asphalt - old asphalt transitions will not be detected Ground

Penetrating Radar (GPR) is used on an ‘as needed’ basis. Applications are the detection of

buried objects, voids under thick concrete slabs, steel presence and depth, locations where

deep investigations are required. It is fairly inexpensive, robust equipment technology and

software is widely available. Deep investigations are possible with low frequency

equipment. Ground Coupled GPR has a speed limitation which is typically less than 10

mph, limited to near surface information, penetration limited in clay material, qualitative

info; usually need expert for interpretation. Air coupled GPR is being used at traffic speed in

combination with surface video imagery. Current limitation is the limited depths of 50 – 60

cm. It is the general feeling that GPR is being oversold.

HMA Infra-red Measurements

Applications are: temperature uniformity of new HMA layers, thermal segregation

detection, creating a permanent log of paving operations - location and duration of paver

stops. Advantages are: segregation of hot mix asphalt a continuing problem, newer lower

cost camera systems, widely available, automated system with 100% coverage, cameras and

guns available but difficult to obtained 100%. Limitations are that equipment not widely

available. Barriers to implementation may be that targets be - given the variability of PG

gradations and mix types, not currently included in specifications

Intelligent Asphalt Compaction

Technology principles: Intelligent Asphalt Compaction for real-time monitoring of

compaction level in the field comprises of a variety of sensors (e.g. accelerometers,

temperature, pressure, displacement) and an onboard computer at the rollers that will

process the signals from the sensors. In combination with GPS, pattern recognition and

classification capability, and accessories including display panels, compaction levels can be

measured in real-time. Applications: quality control/assurance testing of bases and sub-

grades, when to stop rolling/finding weak spots, hot mix applications? Advantages are

100% coverage, related to sub structure support/stiffness. Limitations: what is being

measured is not well understood, limited ability to check individual layers, equipment

availability in USA, typically option when purchasing new rollers. Barriers to

implementation: where should it be used? How can targets be selected? How moisture

content is accounted for?

Rolling Dynamic Deflectometer

The Rolling Dynamic Deflectometer (RDD) is a nondestructive deflection testing device to

measure continuous deflection profiles along highway and airport pavements. It is being

used for the identification of critical sections, cracks, or joints along a pavement test section

that need repair. It is also an effective tool for monitoring deterioration in pavement sections

over time. Research is going on in the USA as well as in The Netherlands5 to improve the

RDD measurements.

Applications: rehabilitation planning on Jointed Concrete Pavements, 100% coverage of load

transfer efficiency. Advantages: faster than other methods. From interviews it was learned

selecting Rehab for Jointed Concrete continues to be a major concern. Limitations: slow

speed (2 kph) and cost. Barriers to implementation: need for a better rolling sensors system,

need to increase speed, improved software required when existing slab has HMA overlay,

influence of environmental conditions needs to be defined.

5 In May 2009 a Danish high speed deflectometer (HSD) has been tested by RWS.


D03031/CG9/0M0/001999/ep ARCADIS 20

Laser Line Scanner

In 2004, a laser based technology was tested for the early detection of ravelling (Eijbersen,

2004). From a practical point of view the first results were not successful, as the laser only

scanned a very limited area (beam size) in the tyre track. Currently, a scanning laser is tested

in a joint effort of TNO and RWS-DID to detect ravelling problems in the pavement surface.

The idea is to mount the laser on a dedicated vehicle to scan the road surface. The scanning

will be done at traffic speed in order to acquire the information with least traffic hindrance.

To measure ravelling successfully the system should fulfil the following two needs: the laser

must scan with a very high density and the accuracy of scanned points should be high.

Whereas the third generation ARAN samples one to six points per meter, the new system

should sample 100 or more points per meter. The scanned points must have a vertical

accuracy in the range of 0,5-1 mm. TNO employs software to simulate different scanner

configurations for ravelling detection. The challenge is to reach the abovementioned relative

high accuracy of less than a millimetre and to distinguish between normal pore spaces in the

asphalt and ravelling. The road will be scanned for every meter and the results will be

interpolated in order to get an idea of the level of ravelling on that stretch of road. If

simulations go well then the aim is to install and test a real laser on a vehicle by the end of

this year.

3.5 OPPORTUNITY I: DETECTION OF PAVEMENT DEFECTS

A better and timely observation of the road surface defects is likely to lower maintenance

cost, support planning schemes and budget allocations and increase safety and comfort of

the drivers. Ravelling and cracking, are today inspected visually on the road or are done ‘in

office’ from ARAN video imagery. Currently, no fully automated technique exists to detect

these defects from video or photos. Cracks can be recognized by a computer successfully,

however, confirmation is still needed by a human expert. For ravelling no (semi-)automated

detection is available at this moment, while at the same time, ravelling is a major issue in the

Netherlands due to the wide spread use of very porous asphalt concrete (Dutch: ‘zeer open

asfalt beton’ or ZOAB). About 80-90% of the Dutch highways are paved with very porous

asphalt concrete. While the advantages are noise reduction and little or no water spray

during rain showers, the open structure of ZOAB makes the asphalt prone to ravelling.

Ravelling is the process where small stones in the top of the asphalt layer are lost, causing

small holes, which lead to more stones loosening, making the holes grow faster over time.

Loose gravel on the highway may cause damage to vehicles, leading to potentially unsafe

situations. In addition, skid resistance increases for ravelled areas, causing faster wear of

tyres and higher noise levels (see also Figure 7).

Experience has learned that in the first five years the most important factor causing ravelling

is traffic (volume and loading). After that period the composition of the asphalt and the

climate (freezing - thawing cycles, UV) are the main causes of ZOAB decline. Other

deterioration features like cracks are typically occurring after the ravelling stage. Rutting

(Dutch: ‘spoorvorming’) hardly occurs in ZOAB layers.


D03031/CG9/0M0/001999/ep ARCADIS 21

Fig.: 7. Left: porous asphalt concrete layer; right: ravelling causing a hole in the pavement.

Important for RWS is to know how much gravel is missing in an area that experiences

ravelling. Currently however, no advanced inspection technique exists for the efficient and

automatic detection of ravelling. Ravelling is inspected visually, while driving slowly along

the hard shoulder, or mapped manually from video imagery. Therefore, an opportunity

would be to use a combination of high resolution video techniques and smart image

analysis algorithms for the automatic detection of these and other defects. The following

techniques have been identified as potential innovations for detection of pavement defects:

‘Smart Dust’, ‘Car as Sensor’, Mobile 3D Video Mapping, Unmanned Aerial Vehicles (UAV),

Spaceborne and Ground Penetrating Radar (GPR).

3.5.1 ‘CAR AS SENSOR’

As can be read in the previous sections, technologies like ARAN are available and provide

accurate assessment of pavement conditions. However, these technologies require specialist

equipment, are dedicated to specific tasks and purchase cost are high. For this reason, a

relatively low number of systems are available and it is not practical to undertake frequent

road network covering inspections, let alone ‘24/7’ monitoring. Inspections of the pavement

condition are therefore done relatively infrequent. RWS currently owns two ARAN vehicles

(one of which is an older version, now used as back-up); it takes two years to survey the

entire Dutch highways network with one vehicle.

Nowadays, the vehicle industry provides standard production vehicles with a wide range of

sensors (see also Annex 3: ‘Car as Sensor’). These sensors are used by systems such as ABS,

Adaptive Cruise Control and Active Suspension (see Figure 8). ‘Car as Sensor’ refers to the

use of these sensors in particular for traffic flow management but also investigations are

going on whether or not these relatively cheap and abundantly available in-car sensor

technologies can collect data about pavement surface conditions. If so, they would both

support the road network management and increase safety and comfort of the users (see

also Wikipedia: Floating Car Data).

http://en.wikipedia.org/wiki/Floating_car_data


D03031/CG9/0M0/001999/ep ARCADIS 22

Fig.: 8. Sensors in modern cars (source: TRL, UK).

During the INTRO (INTRO, 2008a) project various experiments have been carried out with a

variety of sensors that are available in cars (see also Annex 3: ‘Car as Sensor’). It was

demonstrated that technologies that measure vehicle dynamics, such as lateral forces and

wheel speed measurements provide information about surface conditions and irregularities.

A strong correlation was found between lateral forces and irregularities in the longitudinal

profiles. Also a correlation was found between lateral acceleration deviations and rutting.

Much is expected from the vertical acceleration technology that is used in Active Suspension

Control, in order to detect irregularities like potholes or joints. Less successful so far was the

use of ABS and wheel speed data to identify tracks with low skid resistance.

TNO is currently using a special test vehicle that is capable of measuring the following

aspects:

Flatness - sensors on shock absorbers

Roughness - ABS output

Cracks - GPS and Camera

Potholes - GPS and Camera

Unevenness of joints - sensors on shock absorbers

Maximum height of tunnels - detection with laser or infrared

Condition of markings - GPS and camera

Condition of traffic signs - GPS and camera

Condition of information panels - GPS and camera

TNO will use their infrastructure and expertise in the ‘Car as Sensor’ study for RWS6. An

overview of sensor techniques tested by the INTRO (INTRO, 2008b) project and TNO is

provided in Table 4 below.

6 On 22 September this year a meeting will take place between TNO and RWS to define the objectives of

a possible Car as Sensor study.

http://www.fehrl.org/index.php?m=32&mode=download&id_file=5042


D03031/CG9/0M0/001999/ep ARCADIS 23

Table 4: Car technologies vs. road property matrix

The huge amount of information that will be collected, manipulated and disseminated will

become a big challenge for any highway agency and their contractors. They should face this

challenge together with ICT partners, the navigation industry and the automotive industry.

Privacy issues will play a role as the observation will only become valuable when it is linked

with a location. Although the highway agencies like RWS are not interested in who is

sending or receiving the information it should think about how to encourage and implement

public participation. In Israel, the UK and the USA the GreenRoad company installed

vehicle and driver behaviour recording equipment in company car fleets. The employee and

also the fleet manager can learn from the feedback provided by the system. The information

can at the same time be used for deriving anonymous statistics about the road network. In

The Netherlands a similar initiative is currently underway, TomTom car navigation systems

maker and Vodafone mobile communication company work together on real-time traffic

information collection and distribution.

Evaluation

Costs: Because ’Car as Sensor’ technology for measuring road conditions is currently still in

a research and development stage, therefore it is difficult to estimate the costs involved in

the implementation of this technology. To estimate the cost for implementing the car as a

sensor idea is also difficult as it is not one technique, but a combination of existing

technologies. First of all the car sensor data need to be connected to communication and

navigation (GPS) technology. It is possible that development and implementation costs may

well be carried by both car manufacturers and mobile communication and navigation

networks. The interest for the automotive industry would be a safer car, the telecom and car

navigation industry may ask a fee from road users for receiving feedback from the road

network, such as traffic information, weather information, road conditions ahead, and other

info. The cost for data use, processing and retrieval would be for the road network manager.

Benefits: The benefits of ‘Car as Sensor’ are clear; standard production vehicles could

potentially be a cheap, abundant source of condition information for network managers. The

development of in-car sensor technology is still continuing and the introduction in standard

cars and the use of this information by other parties will still take some years. Also the

http://www.greenroad.com/

http://www.tomtom.com/

http://online.vodafone.co.uk/dispatch/Portal/appmanager/vodafone/wrp?_nfpb=true&_pageLabel=templateCClamp&pageID=OS_0089


D03031/CG9/0M0/001999/ep ARCADIS 24

communication technology for data transfer and related privacy issues will slow down the

practical implementation of ‘Car as Sensor’ using standard cars. RWS could start already

using this technology by implementing the most promising technologies in their own fleet

of cars and log the data offline and send the information on a daily basis to the relevant

RWS organisation, the first step to a full scale early warning system. ‘Car as Sensor’ allows a

win-win situation, ‘24/7’ near-real time monitoring of the pavement surface for entire Dutch

highway network and increased serviceability to the road users.

A disadvantage could be that the existing car sensors are not dedicated to the specific task of

road asset monitoring. It is therefore likely that the information gathered by these sensors is

less accurate than the information provided by specialist systems. However, redundancy of

data and intelligent algorithms should allow for the discrimination and filtering of relevant

information. Furthermore, it should be considered to what extent this information will be

used. ‘Car as Sensor’ information should initially be considered as useful to identify road

deterioration in an early stage; hot spot identification. More sophisticated technologies like

mentioned in this report can then be used to exactly measure the extent of the problem. ‘Car

as Sensor’ is most promising to detect the location of pavements defects like ravelling,

potholes and rutting. For ravelling detection, an acoustic sensor may be considered, because

rolling tyre – ravelled pavement contact is noisier than on non-ravelled asphalt.

Safety: Safety seems to be only positively influenced as the output of the car as a sensor

technology is principally focused on the safety of the drivers. If the same technology can

also provide road condition information it will substantially increase both the safety and

comfort of road users.

Contracts: Eventually the technology will have an impact when public engagement is

required. Participatory mapping on this scale has not been tried before in The Netherlands.

Also an impact on data flow management is expected. A pilot with Rijkstwaterstaat’s own

fleet of cars would provide an ideal test ground before introducing the technique in public

vehicles.

Quality: Quality of ‘Car as Sensor’ data will be lower than from dedicated calibrated

sensors, however due to the huge amount of sensors, it is believed that statistical analysis

and data mining techniques can still produce relevant information about highway

conditions.

3.5.2 MOBILE 3D VIDEO MAPPING

Mobile video mapping is specifically designed for recording and measuring features in real-

world coordinates. ‘Video’ is actually misleading, as the system features nine high

resolution (2 megapixel) digital frame (10 fps) cameras that produce non-interlaced frames

for every meter travelled (at 40 kph). The mobile video mapping imagery is geo-referenced,

so every pixel has XYZ information with an error of +/- 5 cm or better. The system features

high resolution digital photo cameras. First tests of the automatic detection of road signs

and other features that are recorded at traffic speed are promising. Recent improvements in

video imagery, global positioning systems combined with inertial measurement units and

parallel data processing enables the mapping of roads and objects provide a level of

accuracy and detail that is unprecedented. Mobile video mapping is done from a vehicle

that has a number of cameras mounted in pairs on the top (see Figure 9). The system

provides 360º, high resolution (see Table 5) geo-referenced stereo imagery for every meter of

road travelled.


D03031/CG9/0M0/001999/ep ARCADIS 25

Table 5: Geometric resolution of the Mobile 3D Video Mapping system

Camera distance to object

(mtrs)Pixel size (cms)

1 0.07

5 0.37

10 0.73

15 1.10

25 1.83

50 3.65

100 7.3

Currently the capacity of the system allows about 300 km of road to be mapped in one day.

All video is geo-referenced using a combination of GPS and INS (Wikipedia: Inertial

Navigation System). Known points in the imagery are used as control points, enabling

location precision of around 4-5 cm.

The Mobile 3D Video Mapping (M3DVM) technique is applied in Belgium for mapping

houses and is currently being tested for road asset mapping. M3DVM is an efficient and safe

technique to collect video data at driving speed with high

location accuracy. The combination of stereo imaging with

GPS-INS positioning makes that pavement conditions can

be stored in a geo-database directly and later retrieved

again for comparison.

A drawback is that currently no good techniques exist for

the automated extraction of defects in road surfaces,

except for cracking. Manual post-processing of video data

will be a major factor in the total cost calculation, until

fully- automated feature detection techniques are able to

perform this task. Only for cracking a number of semi-

automated techniques have been developed

(RoadCrack, WiseCrax, DynaCrack, etc.) that aid the

user in the detection and measurement of cracking

features. However, human intervention is still needed to decide whether or not a feature is a

crack or not. When it comes to safety, since acquisition is done from a car at driving speed

(>= 50 km/h) a matrix sign attached to the rear of the vehicle is needed to warn for a slow

moving car.

Legal and contractual aspects have been investigated; in this case the recorded video

information becomes the property of the user of the system.

ARCADIS is currently testing the 3D measurement capabilities of the system. Further

testing is needed to decide whether or not resolution and position accuracy are sufficient for

highway inspections. Evaluation

Costs: It is expected that the costs for 3D video mapping is comparable with ARAN

operation. The cost-benefit ratio will become better when 3D video data is also used for

structures and objects inspections.

Benefits: Mobile video mapping of pavements could replace on-site visual inspections in the

near future. When the imagery is combined, based on the GPS/INS data, a seamless and

Fig.: 9. Mobile 3D Video Mapping

system

http://en.wikipedia.org/wiki/GPS/INS

http://en.wikipedia.org/wiki/GPS/INS


D03031/CG9/0M0/001999/ep ARCADIS 26

gapless recording of the mapped route is available for in-office inspection. Since images are

geo-referenced, distances can measured on screen and accurate location and size of defects

can be mapped digitally.

Safety: Mobile video mapping is currently done at speeds of 40 kph, hence a warning for a

slow moving vehicle is needed, using a rolling road block, also to keep cars away from the

recording vehicle. An advantage of the relative slow moving vehicle is that passing cars do

not cover frames too long.

Contracts: Different types of contracts are possible. The most advantageous types of

contract are those where the user of the system is also the owner of the recorded data. This

means that the data is free for use and further distribution within an organisation. Some

vendors offer this possibility; while others distribute the data under license, therefore

distribution within an organisation is limited or costly.

Quality: Quality of 3D video mapping is high, a total of nine cameras are used. The location

error of points in the imagery is in the order of 4-5 cm.

3.5.3 UNMANNED AERIAL VEHICLES

A number of companies and organisations have looked or is currently looking into the

possibility to use Unmanned Aerial Vehicles (UAV) for inspections such as: TNO, LARS;

Breijn, tested UAV as part of a PhD project and Miramap, which is described in more detail

below.

The MIRAMAP system uses a helicopter UAV has dimensions of circa 3 meter and exists of

a state-of-the-art navigation and controlling system for autonomous flights from a base

station. The platform is equipped with a live video downlink and an automatic trigger

system for making overlapping digital air photos. Each air photo has position, attitude and

time coding information attached by means of an integrated GPS/IMU system, so Miramap

can deliver up-to-date and very accurate photogrammetric products in a short time span. A

complete workflow was developed that comprises triangulation, 3D mapping on a digital

workstation, automatic DTM calculation and orthophoto production.

The helicopter UAV features mission mode for pre-programmed flight trajectory tracking

that is combined with an automatic camera triggering system of a nadir mounted digital

camera. The digital camera currently installed is a Fuji FinePix S3 Pro with a calibrated 28-

mm lens. Each digital frame counts 12 Mega pixels and comes with position, attitude and

timing information for further photogrammetric processing.

The system enables collection, processing, and delivery of seamless high resolution

orthophoto’s and 3D vector maps to update existing databases. The benefits of this

particular helicopter UAV over traditional aerial survey with large format cameras and/or

terrestrial methodologies are its quick, safe and affordable map updates of areas with the

size of a cross section or highway rest area.

Typical photogrammetric flight parameters and a flight plan of a typical project area are

shown in Table 6 and Figure 10. The flight plan is uploaded to the helicopter UAV that

automatically follows the pre-determined trajectory and triggers the camera at the

appropriate location.

Table 6. Typical Photogrammetric Flight Parameters Flying height 100 m AGL

Side overlap 30%

Forward overlap 60%

Pixel size 2 cm

Single photo coverage 80 by 60 m2

http://www.tno.nl/

http://www.breijn.nl/

http://www.miramap.com/service8/grootschalige%20kartering


D03031/CG9/0M0/001999/ep ARCADIS 27

Fig.: 10. Typical Photogrammetric Flight Plan (Source: Miramap)

The photogrammetric products are generated in a digital workflow of triangulation, block

adjustment, image matching and 3D stereo models. Currently, Ground Control Points

(GCPs) are used to validate the GPS/IMU data of the system. The first results with an

assessment of the product quality are expected to be presented at the workshop.

MIRAMAP and E-Producties have successfully developed and demonstrated a new

helicopter UAV system for the civil market to produce photogrammetric products. The

system is quick, safe and affordable to collect, process and deliver seamless high resolution

orthophoto’s and 3D vector maps of areas with the size of a cross section or highway rest

area. Evaluation

Costs: The cost of the system is the cost of the UAV, cameras and personnel needed.

Benefits: The benefits of a UAV are clear; they can cover large stretches of road in one

operation. Their use can best be described as instruments to make an inventory of road

assets and also partly monitor their conditions (‘hot spot’ detection). In general one can say

that UAVs have a higher operational flexibility then the current airplane and allow the

measuring of elongated tracks. At the same time one should wonder the value adding of

putting optical sensors in the UAV as The Netherlands is annually covered by high

resolution (10 cm) digital airborne photography by CycloMedia.

Safety: An issue for safety would be the fact that drivers could be distracted by the UAV

flying over head.

Contracts: When the technique is developed and acquired by RWS then the contracting

form is not an issue.

Quality: Although image resolution of digital photos acquired from a relatively low altitude

is high, it may not be enough for ravelling or crack detection, therefore the added value to

inspections seems limited.

3.5.4 SPACEBORNE TECHNIQUES

The possibilities of using space borne sensors for road inspections were also investigated.

However, it was soon found that the geometric resolution (0.5-1 m) of satellite imagery do

not match the size of defects (0.01-0.5 m) and layout (long elongated tracks) of highways.

Problems such as cracking and ravelling can therefore not be identified in satellite images.

The only space borne sensor that could possibly contribute to road inspections by measuring

deformations is Synthetic Aperture Radar (SAR; see also Section 4.9.4).

http://www.miramap.com/service8/grootschalige%20kartering

http://www.pegasus4europe.com/pegasus/workshop/documents/contributions/Haarbrink_UAV_full.pdf


D03031/CG9/0M0/001999/ep ARCADIS 28

Evaluation

Costs: Satellite imagery is relatively cheap to acquire, however for the elongated road

networks the square image format of satellite imagery seems not very cost-efficient.

Benefits: No benefits are currently known since image resolutions are not sufficient to

resolve cracking, or ravelling.

Safety: The safety of this technology is not an issue.

Contracts: Unknown.

Quality: Satellite image resolution has improved dramatically over the past decades, from

tens of metres to sub-metre resolution today. Despite the future improvements in resolution,

even a 10 cm pixel size will still not be useful for the inspection requirements of RWS.

3.5.5 GROUND PENETRATING RADAR

Ground-penetrating radar (GPR) is a geophysical method that uses radar pulses to image

the subsurface. This non-destructive method uses electromagnetic radiation in the

microwave band (UHF/VHF frequencies) of the radio spectrum, and detects the reflected

signals from subsurface structures. GPR can be used in a variety of media, including rock,

soil, ice, fresh water, pavements and structures. It can detect objects, changes in material,

and voids and cracks (Source: Wikipedia Ground Penetrating Radar). Ground Penetrating

Radar, or GPR, can be used to discriminate pavement layers transitions through their

different dielectric properties. It is therefore able to detect voids filled with water or air, and

investigate the armour in concrete for example. The greatest advantage of GPR over

traditional coring techniques is that it provides the user with a full cover record of the build

up of subsurface layers. GPR is able to pick up changes in pavement thicknesses as an

indicator for the expected life time of the pavement. Coring will still be needed at those

locations where GPR does show complex transitions. However, the number of cores needed

to understand the subsurface conditions can drastically be reduced.

The most promising systems are the so-called 3D air-coupled GPR systems that employ

multiple antennas mounted on one vehicle. The frequencies used in pavement inspections

are between 150-2000 Mhz, the higher the frequency the lower the penetration depths but

the higher the resolution. For pavements, depths of 1 m are usually sufficient. 3D GPR

provides the user with a lane-wide vertical profile of the subsurface. The recording can be

done at high speed (80 kph), which gives a data density of four samples per meter travelled.

For structure inspections lower speeds with higher sample densities are common.

In one project, 3D GPR was able to locate old concrete pavement layers with a smaller width

than the asphalt concrete layer that was later applied. When coring alone had been done, it

is well possible that the existence of a smaller concrete subsurface layer would have been

missed altogether.

Evaluation

Costs: The cost of 3D GPR will be earned back in a project; the number of cores needed in a

subsurface inspection can be reduced with a factor ten.

Benefits: The clear benefits of high speed GPR techniques are that instead of one core per

100 m, maybe only one per kilometre is needed. Also the lane wide profile gives a lot more

detail than cores can give. GPR can be used for detailed measurements at low speed, e.g. for

structures, but also for network inspections.

Safety: The technique is safe as it is done from a car at driving speed.

http://en.wikipedia.org/wiki/Ground-Penetrating_Radar


D03031/CG9/0M0/001999/ep ARCADIS 29

Contracts: When the technique is developed and acquired by RWS then the contracting

form is not an issue.

Quality: Quality of GPR depends on the wavelength used. As a rule, the shorter the

wavelength used, the higher the resolution. Drawback of higher frequencies is the lower

penetration depth.

3.6 EVALUATION OF OPPORTUNITY I TECHNIQUES

It is expected that the costs for 3D video mapping is comparable with ARAN and other

currently applied technologies. The more as 3D video can also be used for structure

inspections. The initial costs of implementing ‘Car as Sensor’ technologies at a national scale

are expected to be substantial. Not only because of the sensors itself but also because of the

need to set up a proper data flow infrastructure. Initial high costs can be avoided by first

using the RWS fleet and store the information in the car. If a technology appears to be

successful one can expand it towards the commercial and public transport sector. Once

operational, the ‘Car as Sensor’ technology will be very efficient as it will collect road

conditions in a fast and safe way with proper location information (GPS) and increase the

early awareness of pavement defects7. Also, ‘Car as Sensor’ information may be used for

more than just one use.

The benefits of all abovementioned innovative techniques are faster collection of road

conditions with exact location information in a safe way. Also, data may be used for more

than just one use. Video for pavement can also be used for object inventory and inspections.

Positive for the contract or legal side of the video mapping technique is that the user of the

vehicle is also the owner of the recorded data. The technique is mature enough to be applied

in Belgium for mapping housing, and is currently being tested for road assets mapping.

Except for Mobile 3D Video Mapping all innovative techniques are collecting information at

highway traffic speed or are carried out from the air. This means increased safety for the

inspectors as well as less or no hindrance for the road users.

Different types of contracts are possible. The most advantageous types of contract are those

where the user of the system is also the owner of the recorded data. This means that the data

is free for use and further distribution within an organisation. Some vendors offer this

possibility; while others distribute the data under license.

Quality of 3D video mapping will be higher than current (visual) techniques, while car as a

sensor data quality will be lower, which is compensated in part by the frequency and

distribution of data collection.

In Table 7 below the requirements and current inspection techniques are repeated but now

with the addition of where innovations could contribute to inspections.

7 Proper asset management tools are required to take advantage of the early awareness of pavement

defects and optimize the planning to repair or replace pavement.


D03031/CG9/0M0/001999/ep ARCADIS 30

Table 7. Requirements, current techniques and innovations.

Measure /inspect Currently used technique

Car As

Sensor

Space borne

M3DVM UAV GPR

Skid resistance Wheel towed behind car

+

Rutting ARAN-3 laser

+

Transverse and longitudinal profiles

ARAN-3 laser

+

Deflection response FWD Falling weight deflectometer

Construction layers Coring (core samples) +

Cracks (depth) Asphalt cylinders +

Cracks (surface) Visual + +

Ravelling Visual + +

Dirt Visual +

'Local defects' Visual + +

Vegetation Visual +

In order to compare the new techniques described above a scoring method was applied that

is explained in detail in Annex 4. The resulting rank order of techniques is shown in Table 8

below.

Table 8. Ranking of techniques for opportunity I

Automated detection

Rank Speed Cost Traffic Management

Technical Availability and status

Overall

‘Car as Sensor’ 1 Extreme Rapid

Cheap Never Mostly R&D 8.6

Mobile Video mapping

2 Extreme Rapid

Cheap May be required

Mostly Available Not used for pavements

8.2

Unmanned Aerial Vehicle (UAV)

3 Extreme Rapid

Moderate Never Partially R&D 4

Vehicle based GPR

3 Extreme Rapid

Moderate Never Partially Available Not used for pavements

4

Space borne 5

Extreme Rapid

Very Cheap

Never Not Available Not used for pavements

0

3.7 OPPORTUNITY II: PREDICTING PAVEMENT CONDITION TO PREVENT DEFECTS

A second and even more appealing opportunity is using technologies that collect pavement

condition information before physical defects on the surface like ravelling and cracking

occur. Prevention of such defects will help to improve the road conditions and optimize the

priority setting and planning. This section describes innovative techniques that might

contribute in preventing physical defects of the surface layers by measuring specific

pavement properties such as composition, volume concentration of bitumen, type of gravel

used. Hyper-spectral and radiometric fingerprinting techniques were investigated as

possible new techniques and are explained in further detail below.

In addition a study is discussed that uses a scientific approach to reduce the variability in

the process of paving and increase the quality. The latter is of importance among others for

contractors that make agreements with RWS for the production of very porous asphalt

concrete (ZOAB) layers.


D03031/CG9/0M0/001999/ep ARCADIS 31

3.7.1 ‘SMART DUST’

‘Smart Dust’ technology is about the creation of autonomous sensing and communication in

a cubic millimetre application dedicated sensor; a mote. A 'mote' is a tiny sensor node in a

wireless sensor network that is capable of gathering information and communicate with

other connected nodes in the network. Motes can be used to measure light, temperature,

humidity, acceleration, magnetism, movement and other physical conditions. The strength

of using ‘Smart Dust’ technology is the network wide coverage (anywhere), and the ‘24/7’

availability of data (anytime).

Fig.: 11. ‘Smart Dust’ components (source: Berkeley Robotics Lab)

The main components of a sensor node, as shown in Figure 11, are a microcontroller,

transceiver, external memory, power source and one or more sensors (source: Wikipedia).

These sensors connect to each other in the same way that wireless laptops, desktops, and

PDAs around the world are connected to the internet. They require very little power and, as

their price decreases in the coming years, are designed to be disposable. Crossbow

Technologies was the first company to offer motes commercially. Its latest generation of

devices holds a microprocessor, memory, storage, and an internal analogue-to-digital

converter; all integrated into a device roughly the size of a coin. The task is considerably

more complicated than simply beaming readings to some number-crunching computer. For

one thing, the computer would quickly be overwhelmed with data as the number of sensors

increased. For another, the motes would soon run out of battery power if they were

continually radioing data back. The solution is to create intelligent networks of motes that

pre-process the data and only beam back information of interest, such as the stiffness of a

structure, rather than sending back raw data such as real-time sensor readings about how

much each mote has moved.

Placing motes in the pavement during the laying phase and having a complete grid of cheap

sensors would be an appealing challenge. Then, by communicating together, motes could

provide some information about the pavement deformations and deterioration process.

Although the TNO system TISNET (Traffic Infrastructure Sensor NETwork) is principally

focusing on traffic flow management, motes are available that can measure temperature,

moisture, loading, etc. and as such can contribute to continuously monitoring pavement

characteristics. As an example, motes with accelerometers were tested to measure vertical

http://robotics.eecs.berkeley.edu/~pister/SmartDust/

http://en.wikipedia.org/wiki/Motes

http://www.xbow.com/

http://www.xbow.com/

http://www.ieee.tue.nl/symposium/presentations/Sijs%20TISNET_IEEE_symposium_6_4_2009.pdf


D03031/CG9/0M0/001999/ep ARCADIS 32

loading (INTRO, 2008b). Accelerometers8 are used to measure the motion and vibration of a

structure that is exposed. The potential of ‘Smart Dust’ technology is huge but at this

moment the technique is still in a research and development stage. Next to technical

research issues also some practical issues need to be solved such as how and when these

sensors should be put in the asphalt mix, so as not to be destroyed by high paving

temperature. How to supply power once the sensors are in the asphalt layers? When will

manufacturers start to produce these sensors in huge quantities to reduce the price? What

communication infrastructure is needed and how about data management issues? Motes

are successfully implemented though in measuring structure deformation like in one of the

biggest bridges in the world; the Golden Gate bridge in the US (see also Section 4.7.1). An

advantage of using motes as a sensor network opposed to ‘Car as Sensor’ technology is it’s

independency of the automotive industry.

Another promising usage of ‘Smart Dust’ technology for pavement inspections could be the

gathering of traffic information like loading to better understand the relation between traffic

flow and pavement wearing. Also the measuring by motes of some pavement characteristics

like temperature, moisture and electric conductivity can be of interest if such parameters are

indicative for pavement conditions. If in the future more is known about the pavement

composition in relation to degradation processes, for instance bitumen rigidness due to loss

of some chemical components, ‘Smart Dust’ technology could be developed measuring

some chemical components of the pavement and indicate the status of the pavement before

actual defects occur. The tests were not successful as the motes that were being used were

not sufficiently sensitive.

Evaluation

Cost: It was estimated by TNO that merely 130 million motes are needed to cover the entire

Dutch Highway network. It depends on the future price per mote what the total cost of

using ‘smart dust’ in pavements would be.

Benefits: The potential of ‘Smart Dust’ technology is huge but at this moment the technique

is still in a research and development stage. Most promising usage of ‘Smart Dust’

technology for pavement inspections could be the gathering of traffic information like

loading to better understand the relation between traffic flow and pavement wearing. Also

the measuring by motes of some pavement characteristics like temperature, moisture and

electric conductivity can be of interest if such parameters are indicative for pavement

conditions. If in the future more is known about the pavement composition in relation to

degradation processes, for instance bitumen rigidness due to loss of some chemical

components, ‘Smart Dust’ technology could be developed measuring some chemical

components of the pavement and indicate the status of the pavement before actual defects

occur.

Safety: It is not believed that ‘smart dust’ will pose any risk to road users or inspectors.

Contracts: Motes are already successfully implemented in measuring structure

deformations. An advantage of using motes as a sensor network opposed to ‘Car as Sensor’

technology is its independent of the automotive industry.

Quality: Besides the technical research issues also some practical issues still need to be

solved. One is how and when these sensors should be put in the asphalt mix, so as not to be

8 An accelerometer measures the acceleration it experiences relative to freefall. Single- and multi-axis

models are available to detect magnitude and direction of the acceleration as a vector quantity, and can

be used to sense orientation, vibration and shock.

http://www.fehrl.org/index.php?m=32&mode=download&id_file=5042

http://en.wikipedia.org/wiki/Acceleration

http://en.wikipedia.org/wiki/Euclidean_vector


D03031/CG9/0M0/001999/ep ARCADIS 33

destroyed by high paving temperature? How to supply power once the sensors are in the

asphalt layers?

3.7.2 FINGERPRINTING PAVEMENT COMPOSITION

The Netherlands has about 6000 km of highways that are maintained by Rijkswaterstaat

(Dutch Highway Agency). Paving of highways is done by contractors and it is not always

known what materials were or are being used in the asphalt layers. Investigations show

that different lanes on the same stretch of highway can be constructed using different

materials for instance due to the replacement of asphalt for just one lane. Medusa

Explorations is therefore currently validating an innovative technique that uses a gamma

spectrometer and a fingerprinting method to identify the composition of asphalt layers with

driving speed. The gamma spectrometer is placed on pavement engineering and research in

the Netherlands.

Technical Aspects

The idea is to place a gamma spectrometer on a vehicle and measures the unique

combination of 238

U, 234

Th and 40K concentrations of the asphalt and is matched with a

database of asphalt mixes (library). The technology integrates the radiation over a thickness

of circa 30 centimetres. By measuring these isotopes one can provide information about the

gravel being used. By using this technology full lane coverage can be provided about the

pavement composition and allows the reduction of taking cores; For example, a sampling

program needed to find polluted pavement materials (Kempen), which normally would

need 3 cores for every 100 m could now be reduced to one per 1000 km.

The sensor is vehicle based, mounted on a car and can measure the width of lanes at driving

speed (80-120 km/h). Therefore, no lanes have to be closed, and no rolling road block is

needed. The technique allows for collecting data one day and as a rule of thumb the

processing takes another day. GPS (dGPS or RTK/LNK) is used to geo-reference the data,

with the current sensor settings using a frequency of 10Hz a resolution or footprint of circa 3

m is possible. All data is ready for storage in a GIS database. The first results of a test run in

Friesland on the A31 are promising. Benefits of this technology are to provide pavement

composition information in a safe, reliable and non-destructive way for an entire lane at

driving speed. Figure 12 below shows the road layout in latitude and longitude on X and Y.

The colours and numbers indicate the difference in radiation concentrations for different

highway sections caused by different pavement composition.

Fig.: 12. Results of mapping a road with a gamma-spectrometer (source: Medusa Explorations)

1 2

3

4

5

http://www.medusa-online.com/en/


D03031/CG9/0M0/001999/ep ARCADIS 34

Next to measure the composition of existing asphalt layers, another interesting application

of this technique would be to measure the concentrations of isotopes during the

construction of asphalt layers as indicator of the asphalt quality. This would enable the

creation of a national base map of pavement quality. With concurrent measurements it

would, in theory, be possible to measure the wear and tear of asphalt through loss of

material and associated isotopes activity.

Evaluation

Costs: The cost consists of a HPGe detector, vehicle and operational costs to determine the

composition of the pavement.

Benefits: The potential benefits of this technique become greater when more is known about

the wearing processes of pavements. First the complex relations between pavement

composition, traffic volumes, other external factors (e.g. weather) and deterioration rates of

asphalt need to be better understood. The described technology offers a fast and efficient

way to deliver the data that can help predict the remaining lifetimes of a stretch of

pavement. A problem that is currently not solved is that radiation measurements are

integrated over a thickness of circa 30 centimetres, and are not restricted to the upper

surface layer.

Safety: The system is designed to detect pavement compositions at traffic speed, hence

poses no special risk to road users or inspectors.

Contracts: The system was used already in a study of polluted pavements in an area in the

South of Holland called ‚The Kempen‛.

Quality: The vehicle mounted detector will sample lane-wide composition ‘fingerprints’,

integrated over a depth of circa 30 cm.

3.7.3 HYPERSPECTRAL TECHNIQUES

(Airborne) hyper

Hyper-spectral information technology to

characterize road pavement conditions is relatively

new and in an exploratory stage (Herold et al., 2008

and Pascucci et al. , 2008). Imaging spectrometry

basically provides information about the change in

the chemical composition of asphalt; from the

domination of hydrocarbon in the bitumen (new

roads) to domination/increase of the mineral

composition (gravel) in the older roads. It does not

only tell you about the relative age of a road but

more importantly it tells you about the aging ongoing process; the status of hydrocarbon

erosion tells you something about the stiffening of the asphalt and the chances of cracks and

other deterioration processes. The asphalt spectral libraries that are being setup are not only

applicable for airborne surveys but are also useful for terrestrial observations (e.g.

spectrometer sensors on vehicles). To what extent existing libraries can be generically used

is unknown. Placing the sensor on a car instead of an airplane will positively affect the

geometric resolution (airborne geometry is ca. 50 cm). On vehicles one can use an artificial

light source (strobe). With the technology two processes are being measured that cause

opposite spectral reflectance (see Figure 13); the aging of asphalt causes an increase in the

reflectance whereas the increase of cracks (in older roads) causes a decrease in the

reflectance as new underlying bitumen (hydrocarbon) are surfacing. These processes might


D03031/CG9/0M0/001999/ep ARCADIS 35

neutralize the reflectance, making it difficult to distinguish new asphalt from cracked

asphalt layers. Additional ground observations are therefore required to be able to

discriminate between the two processes. Rain and water on the road (absorption) will have

an effect on the reflectance. Also dirt9 and shadowing will have an effect on the measured

reflectance. In Figure 14 below some results of hyper-spectral scanning of healthy and

cracked road surfaces in the USA are shown.

Fig.: 14. Hyper-spectral imaging of pavement

Evaluation

Costs: Hyper-spectrometry for pavement condition analysis is in its exploratory phase, no

dedicated research activities are currently ongoing on this matter. As the hyper-spectral

studies are at the exploratory level, therefore the cost element is not known.

Benefits: Benefits are the hyper-spectral characterization of roads can be done either from

the air or terrestrial, depending on the scale and accuracy required. Hyper-spectral imaging

allows the extrapolation of the retrieved fingerprints over larger stretches and can provide

full road network coverage. Disadvantages that have been found so far are that sample

spectral signatures are required for filling the ‘fingerprint’ database. Not much is known yet

about the relationship between the hyper-spectral signature of the pavement and the

wearing behaviour of the pavement. Predicting pavement wearing on the basis of pavement

composition is therefore not (yet) possible.

Safety: The technique is safe since data acquisition can be done from the air. The technology

might be particularly useful to determine the chemical composition of asphalt and therefore

in estimating the speed of pavement aging. To what extent this type of information is useful

to better map and be able to interfere in a timely manner is currently not known. Would it

be for instance beneficial to replace asphalt if it has reached a certain chemical composition,

just before physical deterioration might start?

9 Unless of course dirt is originating from the pavement itself

Fig.: 13. Aging and reflectance in

time

Fig. 10: Reflectance over time

CHAPTER CHAPTER


D03031/CG9/0M0/001999/ep ARCADIS 36

Contracts: Because the technology is in a R&D stage, intellectual property rights (IPR) or

embedding issues are at present unknown.

Quality: Its practical implementation might be a problem as features like moisture and dirt

affect the measurements. Additional calibration is then necessary. In combination with other

sensors, the use of airborne spectral observations might be useful if large amount of tracks

have to be analyzed. One might conclude that hyper-spectral information can be used to

understand the status of the aging process by means of its chemical composition. However,

at this moment no conclusive remarks can be made about the type of maintenance

measurements that should be made.

3.7.4 ‘A RECIPE FOR GOOD ASPHALT’, ASPARI

The TU Twente Construction Management & Engineering (CM&E) department is currently

doing a study to assess the variability of new asphalt layers during the construction. The

ASPARI (Asphalt Paving Research & Innovation) team consists of three full-time

researchers, as well as a number of graduate students. ASPARI is a co-operative network of

organisations that work together in research projects and technology development to

improve the performance of the asphalt road construction industry. The issues around more

professional approaches to asphalt paving became relevant and pressing due to changes in

the business environment. Some three years ago CM&E researchers started to talk with the

industry about research into the asphalt paving process. The assumption is that the quality

of asphalt road layers is negatively correlated with total variability in the paving process.

The objectives of ASPARI are to reduce variability in all aspects of the pavement process,

from composition and delivery of asphalt to the paving, cooling and compaction of the

asphalt. To that purpose GPS and IR cameras are mounted on supply vehicles, pavers and

rollers to record speed and temperature variations. Also a weather station is setup to record

weather conditions during paving. The measurements are set up so as not to interfere with

the paving process. GPS precision is from 5-10 cm. All information is stored in a GIS

database. The ASPARI founders and co-operating companies have access to this database.

Technical Aspects

ASPARI strives to change the paving industry from a tradition and experience based

process to a scientific, data explicit based process. The goal of ASPARI is to make the paving

process as constant as possible to increase quality, reduce variability and to decrease

premature failures. Benefits are higher quality asphalt of constant quality and with longer

lifetimes in order to support maintenance and better predict and plan pavement renewal. A

GIS record is available with weather information and all information about vehicle

movements and cooling rates. To optimize the paving processes and best practice learning

this information can later be used when failures occur. Confidence building is the major

challenge that ASPARI is facing by introducing the science based method instead of the

traditional experience based method. However, the paving industry seems willing to co-

operate due to new RWS requirements of longer lifetime guarantees. So far the group can

deliver information within one week but new pilots are set up to deliver information in real

time to the contractor’s workforce.

The assumption is that the quality of asphalt road layers is negatively correlated with total

variability in the paving process. Therefore, GPS and IR cameras are mounted on supply

vehicles, pavers and rollers to record speed and temperature variations. Also a weather

station is setup to record weather conditions during paving. The measurements are set up so

as not to interfere with the paving process. GPS precision is from 5-10 cm. All information is

http://www.aspari.utwente.nl/


D03031/CG9/0M0/001999/ep ARCADIS 37

stored in a GIS database. The founders and companies of ASPARI have access to the

database.

Evaluation

Costs: The cost of hardware, IR cameras and GPS is around € 20.000 per vehicle.

Benefits: The benefits are clear; by reducing the variability in various steps of the complete

paving process a more constant and higher quality asphalt layer can be produced. The

combination of GPS, IR cameras and a dedicated GIS database can help constructors and

RWS to make more cost effective maintenance plans and a pavement quality check. The cost

of hardware, IR cameras and GPS is around € 20.000.

Safety: Since the technique is only used during the paving process the risk seems to be

limited as paving speeds are low and remote sensing techniques are applied, enabling

inspectors to stay clear of moving vehicles and potential risk locations.

Contracts: When it comes to maturity, so far the group can deliver data on variations back

within one week. However, a pilot is being setup for later this year to deliver back data in

real time to the contractor’s workforce.

Quality: Quality of the science based method is high.

3.8 EVALUATION OF OPPORTUNITY II TECHNIQUES

For ’’Smart Dust’’ a big factor in the cost calculation is the life time expectancy of sensors.

It is believed that the cost of Medusa will be able to save cost of current inspections that

require the collection of tens of thousands of cores. Medusa will be able to pin-point

problem areas, lowering dramatically the number of cores that need to be taken for further

analysis. Safety of the fingerprinting and hyper-spectral technique is higher, since these are

high-speed techniques. ASPARI is a monitoring during pavement construction and so

should be carried out with care. Quality for Medusa and hyper-spectral techniques are

indicated as neutral, since the alternatives for coring depend completely on good databases

for comparison and identification. When it comes to maturity, Medusa has already been

tested for composition scanning and has been applied in a polluted pavements project for

RWS. The hyper-spectral techniques and the ASPARI programme are still in a pilot stage.

It is assumed by the authors that in the future more will be known about pavement ageing

and deterioration processes. Especially the link between factors such as weather, traffic

volume and individual and total loading on pavement conditions should be studied further.

It was learned that studies are already carried out to learn about pavement behaviour under

stress (Reference: Annex 6: Review & Evaluation Meeting). With calibrated and validated

models that can decrease in skid resistance, rutting, ravelling and cracking to traffic volumes

and load then the abovementioned techniques, can provide some important parameters that

are needed in these models and therefore contribute to the prevention of defects in the

future.




D03031/CG9/0M0/001999/ep ARCADIS 38


Measure /inspect Currently used technique

Smart Dust

Medusa fingerprint

Hyperspectral ASPARI TU

Twente

Skid resistance Wheel towed behind car

+ + + +

Rutting ARAN-3 laser

+ + + +

Transverse and longitudinal profiles

ARAN-3 laser

Deflection response FWD Falling weight deflectometer

+

Construction layers Coring (core samples)

Cracks (depth) Asphalt cylinders

Cracks (surface) Visual + +

Ravelling Visual +

Dirt Visual

'Local defects' Visual

Vegetation Visual

Loose parts Visual


is explained in detail in Annex 4. The resulting rank order of techniques is shown in Table

10 below.

Table 10. Ranking of techniques for opportunity II.

Prevention and prediction



Overall

‘Smart Dust’ 1 Extreme Rapid

Moderate Never Mostly R&D 8

Medusa fingerprint 2 Extreme Rapid

Cheap Never Partially R&D Tested in field

4.3

Hyperspectral 2 Extreme Rapid

Cheap Never Partially R&D 4.3

ASPARI, TU Twente 4 Slow Moderate Never Partially R&D 2.5


D03031/CG9/0M0/001999/ep ARCADIS 39

CHAPTER

4 Structures

4.1 INTRODUCTION

The Netherlands has circa 5000 structures10

(bridges, fly-overs and tunnels). In 1960 the

Netherlands featured 361 km of highways, in 1990 2200 km and in 2007 2400 km. The largest

wave of bridge and tunnel construction was then also needed in the 1960’s and 70’s. With an

average lifetime expectancy of circa 60-80 years, some structures do not longer comply with

current safety standards and needs. One of the biggest challenges that RWS meets has been

the increase in traffic and axle loads of trucks in the last three decades. Nowadays 40-50

metric ton trucks are common while in 1960 8-10 ton trucks were the heaviest vehicles on

the road.

It was found that in fact a relatively small group of structures are in poor technical

condition11

, while most of structures are still in good condition. Most risks are not strictly

condition based, but are caused by loading up to or over theoretical accepted limits.

Experience learns that structural failures are mainly due to human design or construction

errors. Other risks are related to factors unknown during the period of construction. As an

example, about 1200 bridges with a concrete bridge deck do not longer fulfil the current

safety requirements on shear strength due to a flaw in specific design codes in the past.

Incidentally, risks are caused by unknown material properties during design. An example

is the occurrence of damage due to ASR (Alkali Silica Reaction) in concrete. ASR only occurs

if specific aggregates are used in a combination with specific cement types in a humid

environment. For ASR damaged concrete bridges with plate type bridge decks a structural

assessment of the residual bearing capacity needs to be done. After this assessment,

progression of the ASR-damage needs to be monitored. Progress of ASR can best be

monitored by measuring the expansion of the concrete, caused by the ASR. Because the

effects of ASR-expansion are small compared to the effect of temperature and moisture

changes, all tree items need to be monitored simultaneously. By using temperature,

moisture and expansion sensors (for 20 bridges on A59) stakeholders have information on

the development of ASR in these bridges. The bridge decks in highway A59 are equipped

with fully automated monitoring systems that use solar energy, and data transfer is

automated through GPRS communication, and a secure website where data can be analysed

and presented in numbers and graphs showing the ASR expansion over the years. Once

ASR occurs, the risks are mostly related to the initial robustness of a structure. For example

a concrete beam with a 3D reinforcement net is robust, because the reinforcement will still

keep internally ASR damaged concrete in place. On the other hand, a concrete plate type

10 In this report bridges, tunnels and fly-overs are considered as structures (Dutch: ‘kunstwerken’). All

other elements along the highways like gantries are considered as objects (Chapter 5). 11 About 50 structures are in poor condition and were black listed.


D03031/CG9/0M0/001999/ep ARCADIS 40

bridge deck with no shear reinforcement may collapse due to the negative impact of ASR on

the shear strength of affected concrete (see also Section 4.4).

Another issue that was raised during the ‘Expert Meeting’ (Delft, 27 August 2009) is fatigue

in steel bridge constructions. The deterioration is not immediately visible, since it develops

from within the steel, manifesting itself only later with tiny cracks on the surface. This

problem is related strongly to loading of bridges and good vehicle load monitoring could

help predict fatigue in an early stage.

Current experience reveals that visual inspections are adequate as with this method aspects

of deterioration can be timely foreseen (without risks). One might argue that sensor

technologies for structure inspections are only useful on an ad hoc basis once a risk threshold

of a structure has been surpassed. Because of the relatively low cost of prevention and the

mostly expensive sensor solutions the cost benefit ratio of using sensors is high.

4.2 INSPECTION REQUIREMENTS

At this moment three inspections types are relevant to the Dutch structures:

1) Ad hoc structure inspections can be daily (the ‘Schouw’) where the focus is on the

correct functioning of a structure in relation to day-to-day safety issues.

2) The annual condition (Dutch: ‘toestand’) inspection focuses on the condition of the

structures according to NEN norms. Annual structure inspections (condition

inspections) include the visual inspection of paint layers, steel, wood or concrete, and

visible sand bodies. The inspection is mostly done by two persons because of safety

issues. They annotate on notebooks (analogue) and describe the physical features in

rather generic terms. The inspectors are trained not to judge what they see but to

provide an objective description of the structure condition.

3) Every six12

years maintenance (Dutch: ‘instandhouding’) inspections are carried out and

focus on the risk assessment of a structure; RAMS (Reliability, Availability,

Maintenance, Safety). RAMS are risk (severity) scaled according: 0 is negligible risk,

while 5 means an unacceptable risk, and 6 is a calamity (direct closure). This procedure

of risk based inspections was introduced in 2006.

Maintenance decisions for major repairs and replacements and planning are based on the

six-yearly inspections, whereas the regular (small) maintenance is based on standard

procedures and findings in the annual inspections. This means that maintenance is done

when a certain type of wearing causes a visible feature and surpasses a certain risk level.

Whereas the use of the NEN norms do not inform the manager about the risks of a possible

observed deterioration feature, the RAMS inspections tell less about the physical features

that are being observed and more about the risks.

Basically all structure inspections are based on visual observations by experts. To what

extent, sometimes costly, innovative inspection technologies can contribute to structure

inspections should be put in relation to life cycle costs and the expected technical and

functional life span of a structure. A statistical analysis of Dutch civil structures data on

failure rates (van Noordwijk & Klatter, 2004) shows that the expected technical lifespan for

12 Fixed bridges (Dutch: ‘vaste bruggen’) are inspected once every six years, whereas bridges that can

move (Dutch: ‘beweegbare bruggen’) are being examined once every five years.


D03031/CG9/0M0/001999/ep ARCADIS 41

predominantly concrete structures is 80 years. However, if uncertainties regarding future

functional demands are taken into account, a 90% reliability interval shows that the lifespan

may vary from 40 to 100 years. This great range is explained by the fact that under Dutch

circumstances the actual lifespan is dictated by changing functional demands as a result of

environmental developments instead of the technical deterioration of a civil structure. The

relation between offering some of the new inspection technologies should be placed against

the remaining functional and technical life span of a structure and the costs for replacement

or life time extension (Bakker et al., 2006). This holds in particular for the Dutch structures

that are as an average about 50 years old. Another important criterion in this choice is the

economic value of a structure. A list of structures inspections, reasons for inspection and

limitations is given in Table 11 below.

Table 11: Structure inspection requirements and reasons (source: RWS).

Inspection Reasons

Overall quality

Graffiti Offensive

Joints For steel joints - rubber wearing out

Construction

Defamation

Cracks Reinforcement corrosion/corrosion of pre-stressing tendons

ASR (alkali-silica reation) or other chemical degradation

Special reaction within certain concrete mixes.

Surface deterioration Protection of metal from salt and water

Steel/concrete composition Shear connectors failure

Weathering steel

Paint Stop corrosion, need for painting

Hydraulic wear in CSBS

Connections (welds, rivets, bolts) Are weak points in steel construction

Seals

De-icing salts (spray) Salts leak though joints/gaps

Unique risks of specific types of bridges

Construction

Joint seals Can break and 'jump'

Cracks in concrete, rust staining Risk of losing pre-stress

Voids in grouted ducts Tendon corrosion

Fatigue in steel Deformation may reduce load bearing capacity

Cables (look at surface at close range)

Alignment (bending)

Cleanness/comfort

Emergency exits and walkway Safety/emergency

Ventilation Safety/health

Fire resistance Safety/emergency

Lighting Safety/vision

Ground stability Sliding/caving


D03031/CG9/0M0/001999/ep ARCADIS 42


As mentioned earlier the three inspection types mentioned in the previous section are

initially carried out visually (see Table 12) sometimes supported with some photography.

This means that only visible deterioration features will be noticed and possible hidden

defects will remain unnoticed. So far experience learns that invisible deterioration features

do not often reach a critical RAMS level. Based on these inspections and the type of defects,

it is determined if risks may be present that needs further investigation. If that is the case

more in-depth inspections are carried out and may include sensor technologies like the ones

measuring ASR (see also Section 4.4), load testing (e.g. EBK Ingenieur technology13) and

cracks.

According to structure experts the visual inspections are adequate as with this method

deterioration features can be timely foreseen without risks. One might argue that sensor

technologies for structure inspections are only useful on an ad hoc basis once a risk threshold

of a structure has been surpassed. Because of the low costs of prevention and the mostly

expensive sensor solutions the cost benefits ratio of using sensors is high. In Table 12 in

yellow is indicated which inspections are done visually. One can see that a majority of

checks on structures are still visual checks that require visiting the structure.

Table 12: Structure inspection requirements and currently used techniques (source: RWS).

Inspection Currently used techniques

Overall quality Specialist, checklist, visual

ASR (alkali-silica reaction) or other chemical degradation

Sensors on selected bridges

Graffiti Visual

Joints Visual

Construction Visual

Cracks Visual and tests

Surface deterioration Visual

Paint Visual

Hydraulic wear in CSBS* Visual (bulging, deformation, alteration in load or water level)

Joint seals Visual (weekly, expert)

Cracks in concrete, rust staining Visual


Visual

Alignment (bending) Visual

Emergency exits and walkway Check

Fatigue in steel Visual (straightness) / measure (max. deformation)

Connections (welds, rivets, bolts) Look for movement, defects, visual

Weathering steel Visual (irregularities) / thickness measuring

Seals Look for gaps, visual

Deformation Tachymetry and levelling

De-icing salts (spray) Check for high risk areas

Ground stability Geo-technical

13 EBK is using a combination of acoustic and deformation technology for controlled maximum load

testing of structures.

* CSBS: Corrugated steel buried structures

http://www.ebk-ingenieurbuero.de/index.php


D03031/CG9/0M0/001999/ep ARCADIS 43


‘24/7’ ASR monitoring

Since 2001, bridges in the A59 highway in the Netherlands are monitored ‘24/7’ with

sensors (Bakker, J.D., 2004 & 2008). The purpose of these sensors is to measure Alkali-Silica

Reactions (ASR) in concrete. ASR only occurs if specific aggregates are used in a

combination with specific cement types in a humid environment. For ASR damaged

concrete bridges with plate type bridge decks a structural assessment of the residual bearing

capacity needs to be done. After this assessment, progression of the ASR-damage needs to

be monitored. Progress of ASR can best be monitored by measuring the expansion of the

concrete, caused by the ASR. Because the effects of ASR-expansion are small compared to

the effect of temperature and moisture changes, all tree items need to be monitored

simultaneously. By using temperature, moisture and expansion sensors (for 20 bridges on

A59) stakeholders have information on the development of ASR in these bridges. The bridge

decks in highway A59 are equipped with fully automated monitoring systems that use solar

energy, and data transfer is automated through GPRS communication, and a secure website

where data can be analysed and presented in numbers and graphs showing the ASR

expansion over the years. Once ASR occurs, the risks are mostly related to the initial

robustness of a structure. For example a concrete beam with a 3D reinforcement net is

robust, because the reinforcement will still keep internally ASR damaged concrete in place.

On the other hand, a concrete plate type bridge deck with no shear reinforcement may

collapse due to the negative impact of ASR on the shear strength of affected concrete. Access

to sensor data is done through the internet, with a secure and dedicated website that allows

the viewing of archive data and construction of various graphs.

Wind and Structural Health Monitoring System for Bridges in Hong Kong

The advantage of the continuous monitoring of ongoing ASR processes in structures (or

Wind and Structural Health Monitoring System (WASHMS) is a sophisticated bridge

monitoring system, costing US$ 1.3 million, used by the Hong Kong Highways Department

to ensure road user comfort and safety of the Tsing Ma, Ting Kau, and Kap Shui Mun

bridges that run between Hong Kong and the Hong Kong Airport. In order to oversee the

integrity, durability and reliability of the bridges, WASHMS has four different levels of

operation: sensory systems, data acquisition systems, local centralised computer systems

and global central computer system. The sensory system consists of approximately 900

sensors and their relevant interfacing units. With more than 350 sensors on the Tsing Ma

bridge, 350 on Ting Kau and 200 on Kap Shui Mun, the structural behaviour of the bridges is

measured 24 hours a day, seven days a week.

The sensors include accelerometers, strain gauges, displacement transducers, level sensing

stations, anemometers, temperature sensors and dynamic weight-in-motion sensors. They

measure everything from tarmac temperature and strains in structural members to wind

speed and the deflection and rotation of the kilometres of cables and any movement of

processes in the bridge decks and towers. These sensors are the early warning system for the

bridges, providing the essential information that help the Highways Department to

accurately monitor the general) by sensors is that changes in the structure are picked up

immediately.

The structures have been built to withstand up to a one-minute mean wind speed of

95 metres per second. In 1997, when Hong Kong had a direct hit from Typhoon Victor, wind

speeds of 110 to 120 kilometres per hour were recorded. However, the highest wind speed

http://en.wikipedia.org/wiki/Tsing_Ma_Bridge

http://en.wikipedia.org/wiki/Ting_Kau_Bridge

http://en.wikipedia.org/wiki/Kap_Shui_Mun_Bridge

http://en.wikipedia.org/wiki/Hong_Kong

http://en.wikipedia.org/wiki/Hong_Kong_Airport

http://en.wikipedia.org/wiki/Sensory_system

http://en.wikipedia.org/w/index.php?title=Data_acquisition_system&action=edit&redlink=1

http://en.wikipedia.org/wiki/Sensors

http://en.wikipedia.org/wiki/Accelerometer

http://en.wikipedia.org/wiki/Strain_gauge

http://en.wikipedia.org/wiki/Anemometer

http://en.wikipedia.org/wiki/Tarmac

http://en.wikipedia.org/wiki/Deflection

http://en.wikipedia.org/wiki/Rotation

http://en.wikipedia.org/wiki/Cable


D03031/CG9/0M0/001999/ep ARCADIS 44

on record occurred during Typhoon Wanda in 1962 when a 3 second gust wind speed was

recorded at 78.8 metres per second, 284 kilometres per hour.

The information from these hundreds of different sensors is transmitted to the data

acquisition outstation units. There are three data acquisition outstation units on Tsing Ma

Bridge, three on Ting Kau and two on the Kap Shui Mun.

The computing powerhouse for these systems is in the administrative building used by the

Highways Department in Tsing Yi. The local central computer system provides data

collection control, post-processing, transmission and storage. The global system is used for

data acquisition and analysis, assessing the physical conditions and structural functions of

the bridges and for integration and manipulation of the data acquisition, analysis and

assessing processes (Source: Wikipedia). See also Monitoring Hong Kong's Bridges Real-

Time Kinematic Spans The Gap.

Thermography Temperature anomalies in the structure components might be indicative for moisture in the

lining, the presence of cavities or air inclusion are be an indication for inhomogeneous

material composition. In the Netherlands leakages occur in tunnels and viaducts so it might

be useful to technologies that indicate the presence of such phenomena in the structure by

measuring temperature anomalies. Thermography includes the measurement of surface

temperatures of an object. A well-known application of these thermal images is detecting

errors or defects in the thermal insulation of buildings. The thermographic images will show

up weak points in the thermal insulation by way of temperature anomalies. In Figure 15

below an example is given of water leaking in tunnel liking that was discovered by

thermography.

Fig.: 15. Thermal image of water from a leakage in the lining of a tunnel (Source: Spacetec)

When thermographs certain defined temperature and weather conditions should be met.

Therefore, the observation of temperatures inside the structure over a certain period of time

is required before the technology can be applied. Thermal measurements are best made in

winter and at night to avoid the interference of direct sunlight. As the required conditions

are not consistent the date for the measurement - depending on weather - can only be fixed

at short notice. Therefore the measurements must be made quickly and within a short space

of time because of the surface temperatures in the tunnel change permanently. The scanner

systems must be able to cope with certain velocities and speeds. Spacetec is a German

company that applies thermography in an operational way.

http://en.wikipedia.org/wiki/1962_Pacific_typhoon_season#Typhoon_Wanda

http://en.wikipedia.org/wiki/Data_acquisition

http://en.wikipedia.org/wiki/Data_acquisition

http://en.wikipedia.org/wiki/Tsing_Yi

http://en.wikipedia.org/wiki/Post-processing

http://en.wikipedia.org/wiki/Structural_health_monitoring

http://www.gpsworld.com/survey/built-structures/real-time-kinematic-spans-gap-3397

http://www.gpsworld.com/survey/built-structures/real-time-kinematic-spans-gap-3397


D03031/CG9/0M0/001999/ep ARCADIS 45

US Transport Research Bureau

In the US the Transport Research Bureau (TRB) presented the 2nd

Strategic Highway

Research Programme (SHRP2) results in an international conference14

. A number of

structure inspection techniques were identified as leading practice in the US (Source: US

TRB, SHRP2-TTI3). The results of the investigation are summarized below.

Vibration Based Bridge Monitoring

Technology principles: damage detection due to structural degradation of bridges such as

corrosion of rebars, ASR/DEF, determination of fatigue and fracture, damage detection due

to earthquakes, identification of stiffness changes in bridge piers and girders. An advantage

is the technique is a global technique, so only a limited number of sensors is required to

monitor a large bridge. Limitations are maybe that the method is limited because of the data

interpretation. In addition, it is difficult to relate changes in eigenfrequencies and modal

shapes to damage detection since changes in environmental conditions such as temperature

and humidity also cause changes in modal parameters. Also, localization of defects are

difficult since only a few sensors are used and low accuracy due to low frequency.

X-Ray Backscatter

Technology principles: detection of defects in bridge decks, crack detection in pavements,

corrosion of reinforcing steel in concrete, detect material flaws and defects. Advantages:

demonstrated ability to probe through several centimeters of concrete or steel, can provide

images with high resolution and contrast, equipment can be installed in a van and operated

at around 8 kph. Limitations: nuclear technology, limited experience with infrastructure

applications. Barriers to implementation: the aforementioned limitations noted may deter

implementation.

4.5 OPPORTUNITY III: SUPPORTING IN-FIELD INSPECTION TECHNOLOGIES

4.5.1 TAGGING STRUCTURE ELEMENTS

It appears that the condition of the older structures (60s-80s) - and in particular the concrete

- depends on the responsible (sub-)contractor at that time. In the past, concrete was

fabricated by placing a local factory nearby the road construction. This local process resulted

sometimes in poor concrete mixtures and for sure it makes the past construction process and

knowledge about the used material less transparent. Those structures with poor concrete or

cement mixtures are faster deteriorating then structure built by other contractors. Proper

information storage of the structure is a prerequisite for proper asset management.

Information like construction drawings and the construction processes would be very

helpful for the inspections to better understand the risks of a structure. In practice however,

it proves difficult to retrieve such information, especially for older structures. For the

recording, storage and retrieval of RWS structures information the DISK 2006 database is

used. DISK 2006 is not an asset management system by the way. It reveals technical

information of a structure but not the condition of a structure. Conditions of structures are

stores in Excel sheets at district level.

During the construction phase of a structure the objects are labelled in order to know where

they need to be placed. By tagging the objects with information like composition, age, origin

14 2nd International Symposium on Non-Destructive Testing for Design Evaluation and Construction

Inspection, 25 April 2008 Ljubljana, Slovenia

http://www.trb.org/AboutTRB/Public/SHRP2.aspx




D03031/CG9/0M0/001999/ep ARCADIS 46

etc. it would help the inspectors to read out the characteristics of such objects during the

inspections in the field. It would also be helpful in assisting asset managers to better

understand the condition and history of objects within a structure. With the current

development of modern 3D design software tools of Autodesk and Bentley like Navisworks,

Civil3D and Navigator the relation of 3D design, build and maintenance of structures allows

the object approach of structure elements including the incorporation of origin and

maintenance history of these elements. 3D asset management becomes therewith in the

reach of managers (see Figure 16 below).

Fig.: 16. 3D Construction software allows the approach structure items as separate objects and attach

additional information of such objects (attributes).

A simple RFID (Wikipedia: Radio Frequency Identification) tag can also be incorporated

into bridge elements. They can provide the inspector with potentially complicated

information to carry out inspection routines. Finally, having completed an inspection or

operation, the inspector can synchronize the collected data with the rest of their IT system.

In addition, tags can be both read-and-write-capable, and read-only; the write capability

enables information to be written back to the tag from the reader. The system developed and

implemented for one of the pilots was to improve the asset management and maintenance

scheduling system part of the process. The system uses tagging and hand-held devices to

automate asset tracking and the scheduling of maintenance tasks (Source: Innovation &

Research).

The 3D design and construct software approach coincides with the recent 3D inspection

technology developments. Tao (2006) provides an overview of the advantages of the

different 3D information image based (like Mobile 3D Video Mapping, see also Section 3.5.2)

and point cloud based (like Mobile 3D Laser Mapping, see also Section 5.5.1) acquisition

techniques (see also Table 13 below).

Table 13: A comparison of 3D Data Collection Approaches (Tao, 2006).

Performance Image based

(satellite)

Image based

(airborne)

Image based

(ground)

Point cloud-based (airborne)

Point cloud-based (ground)

Accuracy Low Medium High High High

Resolution Low Medium High Medium High

Turn around time

Fast Medium Low High High

Cost Low Medium High High High

Texturing Low Medium High N/A N/A

By taking advantage of the 3D CAD/GIS software development and the progress with 3D

visualization software, inspectors and managers will become able to better follow the

http://usa.autodesk.com/adsk/servlet/index?id=10571060&siteID=123112

http://usa.autodesk.com/adsk/servlet/index?siteID=123112&id=3566722

http://www.bentley.com/nl-NL/Products/Bentley+Navigator/

http://en.wikipedia.org/wiki/RFID

http://www.innovationandresearchfocus.org.uk/articles/html/issue_67/reports_on_radio_frequency_identification.asp

http://www.innovationandresearchfocus.org.uk/articles/html/issue_67/reports_on_radio_frequency_identification.asp


D03031/CG9/0M0/001999/ep ARCADIS 47

conditions of the road features (in particular structures and objects) in time and analyze

defects (including XYZ) in a virtual 3D environment.

Fig.: 17. Vector information as discerned in CAD and GIS systems.

This means that 3D inspection technologies like 3D laser & 3D video empowers asset

management as the observed objects (in terms as being used in GIS applications; see Figure

17) can be better linked (‘tagged’) with condition information.

Fig.: 18. Connections between RFID tag, database and on-site handheld devices.

It provides wirelessly collected data that can be presented through the web browser for

remote access and interface to the existing asset management system. An overview of the

tagging and system is shown in Figure 18 above. The tagging system consists of three

components:

A centralised database with reporting through the web browser.

Wireless hand-held devices with data transfer capability between the database and

RFID tags.

RFID read/write passive tags embedded into or on to the boilers and front doors of

houses in the Bourneville Village.

When inspections are carried out in the open air, then an accurate DGPS and mobile

communications technology may be sufficient, and the physical installation of RFID tags in

structures is not necessary. The idea is that an inspector - carrying a GPS and GPRS enabled

tablet PC or smart phone - reaches a certain structure element, he or she will receive

inspection information based on the inspector’s exact location. The location and relevant

information about the element ready for inspection are stored in a central database. After


D03031/CG9/0M0/001999/ep ARCADIS 48

the inspection is done, inspection results are automatically send back and an update is made

of the database.

Evaluation

Costs: The main conclusion is that this technology is becoming cheaper.

Benefits: RFID tags should offer construction new opportunities to improve the

maintenance of assets and manufacturers to develop smart products and new services. The

potential savings (money and efficiency) that are inspection related are:

Availability of ‘real time’ data.

Reduction in paperwork.

Reducing incidents and associated cost of having incorrect information at the inspection

site.

Web-enabled information system.

Demolition and disposal (end of product life) information.

Safety: Improved health and safety is expected.

Contracts: The technique will help dispute resolution between contractor and RWS. Also it

will allow contractors to offer new value added maintenance services.

Quality: RFID tagging is a proven technique, used in transport logistics, airplane

manufacturing and maintenance; it is used in all modern passports to store bio-metric and

other information. Therefore, it is believed that the technique is mature and reliable.

4.5.2 CABLE INTEGRITY USING ELECTROMAGNETISM

Corrosion of bridge cables or ropes is not always evident from visual inspection, as

corrosion often originates in the interior of cables. Magnetostrictive sensing (MsS)

technology is being developed to identify anomalies in bridge cables from a single location

on each cable. MsS Technology uses structure-borne elastic waves, called guided waves,

which propagate along the structure confined and guided by its geometric boundaries.

Guided waves in relatively low frequencies (under 200 kHz) can propagate a long distance

along the structure at speeds of more than three miles per second. A good example of this is

the sound of the train wheels we can hear from miles away by pressing our ears against a

railroad track. The MsS is a device that generates and detects guided waves

electromagnetically in ferromagnetic materials. With MsS, a pulse of relatively low-

frequency guided waves is launched along a structure from a fixed test location. When the

propagating guided-wave pulse encounters defects, such as corrosion or fatigue damage, a

portion of the waves is reflected back to the original test location, where it can be detected

by the same sensor and analyzed for evidence of structural anomalies. Since these guided

waves propagate at a high speed, MsS Technology can rapidly inspect suspender ropes from

the bridge deck and provide comprehensive structural condition information. Cablsecan™

of Pure Technologies is a vendor that offers such MsS based measuring devices.

Evaluation

Costs: Moderate costs must be expected for this type of inspection.

Benefits: Relative quick availability of ‘real time’ data of invisible potential defects in

tension cables. Inspectors do not have to reach difficult to access locations for visual checks,

but instead can keep their feet on the ground while the system measures where possible

defects are located along cables.

Safety: Inspectors will still need to go to the structure location for checking the state of

cables; hence an element of risk is still present.

http://www.puretechnologiesltd.com/html/cablescan.php


D03031/CG9/0M0/001999/ep ARCADIS 49

Contracts: Services are available world wide and some US bridge owners will outsource this

type of inspection on a regular basis.

Quality: The technique is able to locate where a possible defect is located in a cable.

4.5.3 DIGITAL PHOTOGRAPHY

Bridge inspections are based around visual inspection and the quality and variability of

observations is a concern. In the UK it is currently tested if and how photographic images

can supplement the visual inspections of structures (McRobbie et al, 2007). It is believed that

the use of digital images leads to more consistent observations and the possibility to

quantify developing problems more objectively. Any image collection system to be used in a

real application of this approach must fit the following conditions:

Safe - ideally it should be able to operate from a safe place without the need for a road

closure.

Fast - both to deploy the system in position, and to collect images.

Simple - should only require setting up and starting with no further human input at

any location, other than moving it to the next location.

Reliable - should collect position data (bearing and elevation) for each image repeatedly

and accurately.

High resolution - should collect images at high enough resolution so that they satisfy

the image requirements after the re-projection process.

During the research the image collection and display investigations have been studied of

multiple imaging positions, single imaging positions, spherical images, and mathematically

transforming images to re-project them as if they were taken perpendicular to the face of

interest in order to remove the effects of parallax. The enhanced multiple position collection

set up which was used appears to be the most suitable approach. Currently the method

requires manual intervention to orientate the camera correctly and record distance

measurements. If the original images are of high enough quality and resolution then the

camera orientation data and the distance information can be used to re-project the images

and remove the effect of parallax. This re-projection produces images which are far easier to

interpret manually and are still suitable for use with the various segmentation and

classification algorithms.

Initially the analysis of the photos was performed manually, and the results of this image

based assessment were compared to the onsite assessments of the structure condition. The

first stage of the study has shown that there was sufficient information and detail in the

images to perform meaningful and useful image based condition assessments. The 2nd stage

of the research focused on the use of automatic image processing techniques. Research up

to the end of Stage 2 of this project has found that there is potential for the use of images for

conducting offsite inspections of structures to a level of accuracy that is comparable with

that achieved on-site. Costs-benefits analysis of introducing this technology has not been

performed.

Evaluation

Costs: As it is an extra tool to support the visual inspections it is expected that the costs of

the inspections will become more costly but that the quality and objectivity of the inspection

will increase.


D03031/CG9/0M0/001999/ep ARCADIS 50

Benefits: The support of digital imagery for structure inspections will not only lead to more

objective and consistent observations but is also of importance for its’ archiving value. If

properly archived, the inspector can refer to observations that were made earlier. Image

classification might support the inspector to compare different stages of deterioration and

quantify and qualify the process.

Safety: Inspectors will still need to go to the structure location for taking photographs;

hence an element of risk is always present.

Contracts: Unknown.

Quality: Because of the time consuming process to produce nadir photographs it might be

interesting to test if the Mobile 3D Video Mapping (M3DVM) system as described in Section

3.5.2 offers a good alternative of the digital photography procedure described here.

4.6 EVALUATION OF OPPORTUNITY III TECHNIQUES



14 below.

Table 14. Ranking of techniques for opportunity III.

Aiding inspections Rank Speed Cost Traffic

Management Technical Availability

and status Overall

RFID tagging 1 Rapid Expensive Never Partially Available Not used by RWS

3.2

Cable scan 2 Slow Moderate Never Partially Available Not used by RWS

3

Digital photography 3 Very slow Moderate Never Partially Available Not used by RWS

2.5

4.7 OPPORTUNITY IV: ‘24/7’ MONITORING OF DEFECTS

Bridges are some of the most critical, valuable and highly visible components of the world’s

transportation infrastructure. Proactive monitoring, management and surveillance of these

structures are of the utmost importance to their owners and managers. Monitoring of

structures, seven days a week, 24 hours a day is a possibility using wireless in situ sensors

like ‘motes’. The principles of motes and their possible use in pavement are discussed in

Section 3.7.1. Could the use of motes be of interest to support Structural Health Monitoring

(SHM). SHM is estimating the state of structural health, or detecting the changes in structure

that affect its performance. Two major factors are the time scale of change and the severity

of change. Time scale looks at how quickly the change occurs, and severity is the degree of

change. Two major categories of SHM are disaster response and continuous health

monitoring (ambient vibrations, wind, etc.). Motes focus in particular on the latter through

indirect damage detection by monitoring changes in structural properties and/or behavior).

A number of innovative sensors that are being used or might be of interest for structure

monitoring is provided where the concept of ‘‘Smart Dust’’ or motes and their potential for

pavement monitoring is explained. The same small and more or less dedicated sensors can

also be used in structures to measure: deformation, tension, corrosion, moisture, loads, and

more (see also Wikipedia: List of sensors). Some innovative sensors for structure monitoring

are treated below. These sensors all have their own specific purpose and they often measure

http://en.wikipedia.org/wiki/List_of_sensors


D03031/CG9/0M0/001999/ep ARCADIS 51

just one type of defect. Therefore, it needs to be kept in mind that often a combination of

sensors will be needed to monitor the integrity of a structure.

4.7.1 IN-SITU SENSORS

Golden Gate Bridge Wireless Sensor Network

In 2006 a wireless sensor network (WSN) was designed, installed, and tested on the Golden

Gate Bridge in San Francisco, in order to monitor ambient vibrations (Sukum, 2007). The

motes used were accelerometers for monitoring two directions of movements (vertical and

transverse), and a thermometer to measure accelerometer temperature for compensation.

The cost of each node is about $600. Sixty-four of these motes were distributed over the

main span and southern tower (see Figure 19a and 19b), comprising the largest wireless

vibration sensor network ever installed for structural health monitoring purposes. The

spatially dense array resulted in an increase in effective measurements compared to single,

isolated, sensors, and further allowed high locations to be analyzed easily and accurately.

One of the main challenges that require future investigation is the data sampling and

communication deficits of the sensor network.

Fig.: 19a. Layout of the motes on the Golden Gate bridge Fig.: 19b. One mote

Steel and cable integrity using acoustic technology

In addition to conventional dissolution corrosion (or rusting), high-strength steels are

susceptible to failure through brittle fracture caused by stress corrosion, hydrogen

embrittlement and fatigue. These corrosion mechanisms cause a significant loss of ductility

in the steel, and failure can occur without measuring a gradual loss of cross-section in the

wire. The presence of corrosion in high-strength steel wire in bridges can have serious

consequences and eventual can lead to the collapse of a bridge. In most cases, corrosion of

high-strength steel wire in bridges is not visually evident. Most of the methods used to

investigate the condition of stressed wires in post-tensioned concrete structures and bridges

involve a representative sampling of the overall structure. Hence, there is a possibility that

corrosion and wire failure in localized areas may not be detected. A comprehensive health

monitoring system that can provide information on deterioration for the entire structure

would therefore be useful. A continuous acoustic health monitoring system developed for

this purpose has been applied to various bridges. Such an acoustic monitoring system has

been developed by Pure Technologies called SoundPrint®. The system has been developed

from the observation that failures of pre-stressing wires generated an audible acoustic

response. If these events had frequency or energy characteristics sufficiently different from

ambient acoustic activity in a structure, it would be possible to identify the events, as well as

their location and time of occurrence. The types of sensors used were broadband piezo-

electric accelerometers. Piezo-electric crystals are man-made or naturally occurring crystals

that produce a charge output when they are compressed, flexed or subjected to shear forces.

http://puretechnologiesltd.com/html/bridges.php


D03031/CG9/0M0/001999/ep ARCADIS 52

The tool helps to ensure the long term integrity of post-tensioned, pre-tensioned and cable

stayed bridges. The so-called SoundPrint® sensors detect and locate tension wire failures

through continuous non-intrusive monitoring. By analyzing the time taken by the energy

wave caused by the break as it travelled through the concrete to arrive at different sensors,

the software was able to calculate the location of the wire break, usually to within 300 - 600

mm of the actual location. The capability of the system to accurately identify and locate wire

break events showed that the system to be 100% correct when spontaneous events classified

as "probable wire breaks" were investigated versus "non-wire break events".

Evaluation

Costs: Most sensors are expensive calibrated devices.

Benefits: ‘24/7’ availability of ‘real time’ data of potentially invisible defects ‘in the heart’ of

structures. Dedicated sensors are developed for specific tasks, whether it is measuring

pressure, moisture, temperatures or crack widths. Online monitoring of sensors enables an

early warning system when using software that checks pre-set thresholds against sensor

readings. Measurements are archived and therefore can readily be used for evaluation and

analysis.

Safety: Remote ‘online’ monitoring is possible; therefore no safety issues for road users or

inspectors are to be expected once the sensors are in place.

Contracts: Long time service and maintenance contracts needed, comparable to A59 ASR

monitoring contracts.

Quality: Calibrated sensors deliver very accurate readings of the parameter they measure.

One prerequisite is that the sensor is measuring what the inspector wants it to measure. In

one example a sensor was measuring moisture in the shafts that was drilled rather than the

moisture in the surrounding concrete. The latter was solved by a decrease in drilled shaft

diameters.

4.8 EVALUATION OF OPPORTUNITY IV TECHNIQUES




Measure /inspect Currently used techniques

In-situ sensors

Overall quality Specialist checklist


Sensors on selected bridges +

Ground stability Geo-technics

Graffiti Visual

Joints Visual +

Construction Visual

Cracks Visual and tests +


Paint Visual +

Hydraulic wear in CSBS* Visual (bulging, deformation, alteration in load or water level)

Joint seals [voegen] Visual (weekly, expert) +

Cracks in concrete, rust staining Visual +

Cables (look at surface at close range) Visual +

Alignment (bending) Tachymetry and levelling +



D03031/CG9/0M0/001999/ep ARCADIS 53

Measure /inspect Currently used techniques

In-situ sensors

Fatigue in steel visual (straightness) measure (max. deformation)

+

Connections (welds, rivets, bolts) Look for movement, defects

Weathering steel* Visual (irregularities) Measuring thickness

+

Seals Look for gaps +


De-icing salts (spray) Check for high risk areas +


is explained in detail in Annex 4. The resulting rank order of techniques is shown in

Table 16 below. The costs of sensors are indicated as ‘moderate’. This indication also

includes the positioning of sensors in the structure, calibrating them, maintaining them,

setting up a wireless network (if applicable) and storing and interpreting the data readings.

This makes the initial costs high but at the end the benefits may prevail as less visual

inspections are required, timely identification of wearing might safe substantial

maintenance costs, less traffic hindrance which has a huge economic effect and - last but not

least - the safety of inspectors increases. For each of the above technologies (and for

structure inspection technologies at large) holds that a sound decision should be made about

the expected life cycle costs when implementing a (semi-)automatic technology in relation

to the expected life span and economic value of a structure

Table 16. Ranking of techniques for opportunity IV.

Automated detection



Overall

In-situ sensors

1 Extreme Rapid

Moderate Never Mostly Available Dedicated sensors are used by RWS

6.8

4.9 OPPORTUNITY II: PREVENTION OF DEFECTS

In this section techniques are described that help prevent corrosion of steel or cracking due

to deformations. The latter, deformation monitoring can be required for the following

applications relevant to RWS and HA:

Roads

Tunnels

Bridges and Viaducts

Foundations

Construction sites

Deformation monitoring (also referred to as Deformation survey) is the systematic

measurement and tracking of the alteration in the shape or dimensions of an object as a

result of the application of stress to it. Deformation monitoring is primarily related to the

field of applied surveying, but may also be related to the civil engineering, mechanical

engineering, plant construction, soil and rock stability mechanics. The causes for

deformation monitoring are changes in the bedrock, increase or decrease of weight, changes

of the material properties or outside influences (see also Wikipedia: Deformation

Monitoring). Early detection of deteriorating protective paint layers or small deformations

can help prevent larger problems, such as corrosion and cracking.

http://en.wikipedia.org/wiki/Deformation

http://en.wikipedia.org/wiki/Deformation_monitoring

http://en.wikipedia.org/wiki/Deformation_monitoring


D03031/CG9/0M0/001999/ep ARCADIS 54

4.9.1 SMART PAINT – C-CUBE

The status of the building material and the construction part is difficult to measure.

Moreover so as certain ageing and deterioration processes are not always well-known. The

use of osmosis meters is an example of measuring the condition of the concrete. TNO/TU

Delft in a joint project developed the C-cube technology to measure the ageing of paint. It

was successful, however, it only measures paint conditions on steel, not on other surfaces

such as concrete or wood.

Bridges, buildings, locks are constructions whose sustainability is ensured by a protective

layer coating. Coatings protect steel structures against corrosion by forming a barrier

between the metal and its surroundings. C-cube provides a coating where the barrier

function can be measured enabling prediction of when coated steel will start to rust. Some

coatings are better then other coatings and after a while the application of a new layer

coating is necessary. It will cost an organization hundreds of thousands of Euros when it is

decided to do this kind of maintenance. The company C-Cube International offers managers

detailed information about the quality of the coating, to support such decisions. The

dilemma often is, when painting too early money is wasted, but when painting too late rust

will form and the construction will last less long."

The quality of coating is often measured by a visual scan, which hardly explains how

corrosion develops. The quality of paint layers deteriorates once applied, even before

rusting occurs. The thin layer of paint changes over time from an impermeable barrier into a

kind of sponge that let water and salt pass through. A problem is that this is often not

immediately visible.

C-Cube Smart Paint will give information in numbers and graphs on how the quality of the

paint is. Public and private parties have already expressed interest in the C-Cube

International developed method. To have the ability to predict when corrosion will occur

enables: timely scheduling of maintenance, the possibility to determine how maintenance is

performed, and checking whether a coating is correctly applied.

Evaluation

Costs: Moderate.

Benefits: ‘Smart’ paint technology developments like C-cube is ongoing and should be

closely followed by RWS. So far, the C-cube technology only measures paint conditions on

steel, not on other surfaces such as concrete or wood. At the moment C-Cube exists as a

prototype in the basement and needs further adaptation to work in the field. As an example

the interpretation and signalling of the condition of the coatings needs further exploration.

Contract: Commercially available product.

Safety: Not an issue

Quality: For maintenance more knowledge is required to better understand the relation

between the dielectric readings and the wearing process to be able to decide when coatings

need to be replaced.

4.9.2 FIBRE OPTICS

Geo-detect® is a textile with embedded fibre optic sensor technology. The sensor warns in a

very early stage for unwanted deformations. The geo-textiles are manufactured with optical

http://www.tudelft.nl/live/pagina.jsp?id=263495dc-791d-4257-9c4e-d57e40d4d008&lang=en&binary=/doc/DO-07-4-2smartaspaint.pdf

http://www.yesdelft.nl/IncubationCentre/IncubationCentreNieuws/tabid/85/YearMonth/200808/ItemID/63/Title/C-Cube%20International%20bespaart%20onderhoudskosten/Default.aspx

http://www.inventec.nl/english/products/Geodetect.pdf


D03031/CG9/0M0/001999/ep ARCADIS 55

glass fibres woven in. These glass fibres feature at selected distances so called Fibre Bragg

Gratings (FBG's). An FBG is a location on the glass fibre where the refractive index of the

fibre material has been modified over a short length (3 - 10mm) by a laser. The key is that for

laser light that is send through the glass fibre, only a specific wavelength is reflected. If a

length change occurs at an FBG then the wavelength of the reflected light will also change.

This change is measured by a scanner and translated into the fibre length change and the

movement or deformation of the body in which the fibre is located (see Figure 20 below).

Fig.: 20. Function of an FBG (source: Geo-detect).

With multiple locations along the same fibre treated differently by laser different FBG’s are

created with their own specific wavelength rebound. In this way the location in the

construction can also be identified. By applying a particular pattern of FBG's, the distributed

form change behaviour over the entire surface of the geo-textile is made visible. The

detection pattern depends on the distribution of the fibres, the distance between FBG ‘s and

the degree of variation in FBG. For each individual application, the configuration can be

adapted to the requirements of the client. The system can be equipped with alarm and

preventive alarm functions that warn when preset values are reached (see Figure 21).

Fig.: 21. Shape change of geo-detect textile caused by a crack (Source: Geo-detect).

A scanner is permanently installed in a suitable location (up to 10km distance of the

measured object). The measurements are stored and can be viewed on a remote computer in

real-time via a wireless connection. Any shape changes are detected with a handheld

monitor. This very cost-effective implementation is ideal for e.g. dike monitoring.




D03031/CG9/0M0/001999/ep ARCADIS 56

Evaluation

Costs: Expensive

Benefits: Very sensitive technique to measure deformation, temperature and strain. Using

FBG’s, problem locations can be accurately identified. The sensitivity of the system enables

early warning.

Contract: Commercially available in Netherlands, however still also R&D. Was tested in the

IJkdijk project (Wikipedia: IJkdijk).

Safety: Not an issue, and traffic management is not required.




4.9.3 DEFORMATION SENSORS

Electronic levelling sensors can be used to measure ‘24/7’ deformations in buildings and

structures. StabiAlert is such a system where every five seconds a reading is registered by

the electronic levelling sensors, which are stored and can be remotely followed.

Many companies install and/or produce electronic settlement sensors that can measure

deformations in constructions due to overloading, nearby earth moving works or

movements in soil through piping or other geo-physical processes. The electronic level

sensors could also be placed at strategic points in bridges to continuously monitor

deformations of structures. Evaluation

Costs: Moderate, depending on life time expectancy of the structure.

Benefits: Very sensitive technique to measure deformations. The sensitivity of the system

enables early warning.

Contract: Commercially available in Netherlands.

Safety: Not an issue, and traffic management is not required.




4.9.4 INSAR TECHNOLOGY

Most SAR applications make use of the amplitude of the return signal, and ignore the phase

data. Synthetic Aperture Radars (SAR) produce all weather, day and night, high resolution

images of the Earth's surface providing useful information about the physical characteristics

of the ground and of the vegetation canopy, such as surface roughness, soil moisture, tree

height and bio-mass estimates. Most of those SAR applications make use of the amplitude of

the return signal, and ignore the phase data. However SAR interferometry (Wikipedia:

InSAR) uses the phase of the reflected radiation. InSAR technology combines two or more

SAR images of the same area and using algorithms to calculate differences in the phase of

the waves returning to the satellite. Since the outgoing wave is produced by the satellite, the

phase is known, and can be compared to the phase of the return signal. In principle the

surface displacement measurement can have a precision of 2%-5% of the SAR wavelength.

Typical SAR wavelengths are in the range 3-30 cm, implying millimeter to centimeter

precision in surface (see also Figure 22 below). It has applications for geophysical

http://en.wikipedia.org/wiki/IJkdijk

http://www.stabialert.nl/index.php?id=3

http://en.wikipedia.org/wiki/Amplitude

http://en.wikipedia.org/wiki/Phase_(waves)

http://en.wikipedia.org/wiki/Amplitude

http://en.wikipedia.org/wiki/Phase_(waves)

http://www-radar.jpl.nasa.gov/sect323/InSar4crust/SarInterferometry.html

http://en.wikipedia.org/wiki/Interferometric_SAR


D03031/CG9/0M0/001999/ep ARCADIS 57

monitoring of natural hazards, for example earthquakes, volcanoes and landslides, and also

in structural engineering, in particular monitoring of subsidence and structural stability.

The phase of the return wave depends on the distance to the ground, since the path length

to the ground and back will consist of a number of whole wavelengths plus some fraction of

a wavelength. This is observable as a phase difference or phase shift in the returning wave.

The total distance to the satellite (i.e. the number of whole wavelengths) is not known, but

the extra fraction of a wavelength can be measured extremely accurately.

Fig.: 22. Deformation study of the Kornwerderzand structure in the period 1992-2002 based on ERS images. The blue points at the east side of the bridge indicate a relative displacement of ca. 3 mm/year.

A closer field study shows indeed deformation cracks with ca. 5 cm vertical displacement (source: HansjeBrinker.net)

InSAR is of particular interest to monitor relative deformation of manmade structures that

avail of rectangular corners; so called permanent scatterers. Although one can not exactly

indicate what part of a structure causes the reflection, it is not always known what is the

permanent scatterer, however, InSAR has proven to be a useful technology for hotspot

indentification of structure deformation over large areas. This can be done on a monthly

basis.

SAR data availability

A variety of factors govern the choice of images which can be used for interferometry. The

simplest is data availability - radar instruments used for interferometry commonly do not

operate continuously, but acquire data only when programmed to do so. For future

requirements it may be possible to program SAR data acquisition, but for many areas of the

world archived data may be sparse. This however is not the case for The Netherlands where

SAR data was recorded as early as 1992. This makes the use of InSAR technology in the

Netherlands of particular interest as the archived images allow studying the deformation

history of for instance road structures. Data availability is further constrained by baseline

criteria. The two images or stack of images if one wants to study deformation in a larger

time span, must be accurately co-registered to a sub-pixel level to ensure that the same

http://en.wikipedia.org/wiki/Wavelengths

http://en.wikipedia.org/wiki/Phase_(waves)#Phase_difference

http://en.wikipedia.org/wiki/Image_registration


D03031/CG9/0M0/001999/ep ARCADIS 58

ground targets are contributing to that pixel. Where co-registration is poor or the maximum

baseline is exceeded the pixel phase will become incoherent - the phase becomes essentially

random from pixel to pixel rather than varying smoothly, and the area appears noisy. This is

also true for anything else that changes the contributions to the phase within each pixel, for

example changes to the ground targets in each pixel caused by vegetation growth,

landslides, agriculture or snow cover. Although atmospheric corrections are necessary, in

general it can be stated that weather conditions do not affect.

Specialist work

Although professional commercial-off-the-shelf (COTS) software is available it is strongly

recommended that InSAR specialist companies are hired to perform the tasks and provide

the highway agency with point deformation information. Experience learns that the

selection of SAR data in itself is already a very critical element in the processing path that

determines the final results.

ALOS

Most used SAR data are ERS and Radarsat. Recent InSAR studies are also looking at the

potential of ALOS data. ALOS-PALSAR is from the Japanese Space Agency and has a R&D

focus. Based on the data availability and required accuracy one should consider what type

of SAR data are required. Whereas ERS data from the European Space Agency can reveal

important information for the Netherlands, Radarsat might provide better operational

services.

Evaluation

Costs: Archived SAR data sets are mostly cheaper then images that need to be acquired on

request (see also Table 17 below). As the SAR imagery covers a large area it will probably

cover a substantial amount of structures. This makes that the InSAR technology per

structure observation might become relatively cheap.

Benefits: InSAR technology is particular useful as hot spot indicator of relative

deformations of road structures over large areas. By using archived data sets a first

indication of the amount of scatterers of a structure (Step 1) will make it clear if this

technique is useful as hot spot indicator of structure deformation.

Table 17. Pricing of SAR data

SAR data type Costs (€) FOV (km

2) Resolution (m)

Archive New

ERS 400.- 700.- 100x100 30

Radarsat-1 standard 930.- 2350.- 100x100 25

Radarsat-1 fine 930.- 2350.- 50x50 8

Radarsat-2 standard -* 2400.- 100x100 25

Radarsat-2 ultra fine - 3350.- 20x20 3

ALOS-PALSAR polarimetric

620.- 30x30 30

ALOS- PALSAR fine 620.- 70x70 20

*No archive data policy yet available as Radarsat-2 has been active since early 2008

Safety: Not an issue for the space borne technology.

Contracts: The distributing space agency will remain owner of the raw satellite imagery.

This means that purchaser of the images is not allowed to further distribute the images

except with written approval of the agency or its reseller. Dependent on the agreement


D03031/CG9/0M0/001999/ep ARCADIS 59

between the SAR value adding organization and the customer, the customer can obtain the

ownership of the derived information.

Quality: It should be considered that the more data sets are available for a certain area, the

more reliable are the results. This can mean that 10-20 scenes per area (if available) are

required to provide reliable information. Most probably, the standard images reveal

sufficient information to identify hotspots. As one SAR scene can cover an area up to

100x100 sq km (field of view) one should consider the costs to identify hotspots with InSAR

technology on the basis of the amount of structures that can be identified by one SAR scene.

Next to the data costs one should consider the expert (processing) costs. Activities that need

to be considered by the expert as well as the highway agency are:

Step 1: Are the structures providing sufficient and reliable permanent scatterers?

Step 2: What is the deformation of these scatterers in time?

Step 3: Regular (monthly?) provision of deformation parameters?

The information can be provided in a digital format through Internet and linked with the

structures that are present in the respective databases.

4.10 EVALUATION OF OPPORTUNITY V TECHNOLOGIES



Table 18. Requirements, current techniques and possible innovations.


Smart Paint

Deformation Sensors

Fibre Optics

InSAR

Overall quality Specialist checklist

Graffiti Visual

Joints Visual +

Construction Visual

Cracks Visual and tests +


Paint Visual +

Hydraulic wear in CSBS Visual (bulging, deformation, alteration in load or water level)

+ +

Joint seals [voegen] Visual (weekly, expert)

+

Cracks in concrete, rust staining Visual


Visual*

Alignment (bending) Tachymetry and levelling

+ +


Fatigue in steel visual (straightness) measure (max. deformation)

Connections (welds, rivets, bolts) Look for movement, defects

Weathering steel Visual (irregularities) Measuring thickness

Seals Look for gaps


+ + +

De-icing salts (spray) Check for high risk areas


Sensors on selected bridges


D03031/CG9/0M0/001999/ep ARCADIS 60


Smart Paint

Deformation Sensors

Fibre Optics

InSAR

Steel/concrete composition Separation between flange and concrete?

Ground stability Geo-technics +



19 below.

Table 19. Ranking of techniques for opportunity V.

Prevention and prediction



Overall

InSAR 1 Extremely rapid

Cheap Never Partially R&D 4.7

Fibre optics – FBG’s 2 Extremely rapid

Expensive Never Partially Available Not in RWS

4.1

Deformation sensor 2 Extremely rapid

Expensive Never Partially Available Not in RWS

4.1

Smart paint 4 Moderate Moderate Never Partially Available Not in RWS

3.4


D03031/CG9/0M0/001999/ep ARCADIS 61

CHAPTER

5 Objects

5.1 INTRODUCTION

In this chapter the inspections of some other relevant road objects are discussed that are

needed to ensure the safe and correct use of roads. All other assets than the ones described

under pavements and structures are categorized as objects. In this chapter we treat the

inspection of safety barriers, road signs, lighting, gantries for the support of signs, and white

lines.

5.2 INSPECTION REQUIREMENT

A variety of ‘defects’ can occur regarding the road objects. From to dense of to high

vegetation along the roads hampering safe drive conditions to rust on the safety barriers

that might weaken its’ actual functions. Whereas DVM functionality is controlled

electronically or by means of software (see Section 1.3), the condition of the physical

structures like gantries is checked visually (see Table 20). Inspections that might support

Table 20. Categories of object categories and inspections (source: RWS).

Categories Inspections

Safety barriers Gaps

Broken, deformed of cracked components

Structure – nuts / bolts / welds

Corrosion (reinforcement)

Tension

Incorrect height

Anchorages (i.e. posts) - movement and ground condition

Ingress of water to post sockets

Excessive weeds

Concrete cracks/spalling

Gantries Welding

Paint damage/rust

Foundation failure

Height

Signage Readability/visibility

White lines Reflectivity

Vegetation Length


D03031/CG9/0M0/001999/ep ARCADIS 62


As can be seen in Table 21 below, inspections of road objects are still done visually, either by

stopping by or while slowly driving by an object. Sometimes more detailed inspections are

needed using specialized inspection tools. It is believed that a big improvement would be to

use a video or laser technique to record the position and objects accurately.

Table 21: Object inspection requirements and currently used techniques (source: RWS-DVS).

Category Inspect Technique currently used

Safety barriers

Alignment and height Visual, drive by

Gantries

Welding Visual, drive by, X-ray

Paint damage/rust Visual

Foundation failure visual, vertical alignment

Signage Readability/visibility Visual, drive by

White lines Reflectivity Visual, drive by

Vegetation Length Visual, drive by

Road marking

Reflectance Luminance factor and colour; TEST 95

Discoloration and reduction in the luminance factor

Luminance factor and colour; TEST 95

Fencing Condition visual

It is believed that a big improvement would be to use video and or laser techniques to

record the state and position of objects accurately.


Belgium uses a Mobile 3D Video Mapping technique to map road signs and other objects in

urban areas.

Visual Roads Database

To support the management and maintenance of roads, a ‘Visual Roads Database’ has large

potential. The number of field survey can be reduced significantly. Furthermore, the

database can be used for the maintenance and checking of other data sets of road

information. The quality of operation increases, road safety is improved and working with

the Visual Roads Database is a joyful experience. These are the results from an investigation

of the Directorate-General for Public Works and Water Management and market partners

(Gerke and Vosselman, 2008). A brief cost-benefit analysis showed that the total cost (collect

video, develop software, post-processing of images) were higher than when using the classic

way of visually inspecting.

Using this method only for maintaining a database would therefore not be cost-saving.

However, once the database is filled with images, they can also be used for other purposes.

Therefore, it was decided to do a broader study and find what other tasks and activities

would benefit from a Visual Roads Database (VWD) of the Rijkswaterstaat roads network

(Geo-Info, 2008-6).

Since post-processing by people is still costly and tedious work, a lot would be gained when

objects could be detected automatically in imagery. This would dramatically decrease the

number of video stills that now have to be processed manually during ‘on-screen’

inspections (see also Figure 23 below).


D03031/CG9/0M0/001999/ep ARCADIS 63

Fig.: 23. On-screen inspection of defects.

5.5 OPPORTUNITY VI: AUTOMATED INSPECTIONS

The ideal inspections of objects would comprise the recording of objects at driving speed

and evaluation of defects by a computer. This may still sound like a distant future, however,

recent developments in camera resolution and computer storage and machine vision bring

this scenario a little closer again.

5.5.1 MOBILE 3D LASER MAPPING

Mobile 3D Laser Mapping (M3DLM) system is based on a set of 360° laser scanners

positioned on a moving platform, in this case a van. Lidar technology15

is made for Mobile

Mapping because lidar can maintain high accuracy and high resolution from rapidly

moving platforms. They operate at very rapid measurement rates (up to 200.000 pulses a

second) to produce inherently 3-dimensional data: Fast, accurate, eye-safe lasers coupled to

high-speed scanners with large fields of view, operating day or night to collect survey-grade

measurements. M3DLM is a high speed inspection technique that normally will not require

the closure of lanes as the gathering of information is at traffic speed. With the current

configuration an area of about 100 meter wide can be recorded in one instance meaning that

several lanes can be recorded instantaneously. Sometimes a second or third scan is required

for higher accuracy and point density. Speeds during scanning range from 40-120 km/h but

is typically 80 km/h. A combination of GPS and an inertial navigation system (INS) is used

to correct for movements of the vehicle carrying the laser and frame cameras (see also

Figure 24). The X, Y, Z-error in point locations is not more than 2,5 cm at a speed of

40 km/h. At a driving speed of 80 km/h the location accuracy is circa 4-5 cm, which is

deemed sufficient for various inspection standards.

15 LIDAR (Light Detection and Ranging) is an optical remote sensing technology that measures

properties of scattered light to find range and/or other information of a distant target. The prevalent

method to determine distance to an object or surface is to use laser pulses.


D03031/CG9/0M0/001999/ep ARCADIS 64

Fig.: 24: Vehicle position and movement is accurately measured by GPS/INS.

In addition video cameras are being used in the M3DLM configuration to support laser data

interpretation (Figure 25).

Fig.: 25. Each laser pixel is assigned with a colour coming from the video imagery to support laser

data interpretation.

The distance inspected each day is limited to 80 km; this is mainly due to the large data

volumes generated by the system. Examples of a test drive in the Netherlands can be found

in the Figures 26 and 27.


D03031/CG9/0M0/001999/ep ARCADIS 65

Fig.: 26. Safety barrier, white lines and road signs in a 3D laser scan.

Fig.: 27 Laser scanning result of a gantry.

Cost of data collection is approximately € 100 per kilometre, while the cost of man-hours for

post-processing of data is two days of work per one day scanning. The system can be rented

or purchased. Recorded data is property of the user of the system, whether it is rented or

owned by the agency. Cost of purchasing the new hardware is relatively high, with 1.3 M€

for a complete system, vehicle, lasers and cameras, and including the necessary software.

In the Netherlands, M3DLM is currently being introduced for the retrieval of XYZ values of

structures. It is not being used yet for inspections of objects. In combination with 3D video

imagery it might be a powerful instrument to inspect the geometric characteristics of objects.

Benefits:

Fast recording of road, structures and terrain.

Safe laser for the recording team and other road users

The hindrance of traffic is minimal.

The accuracy and reliability are comparable with total station measurements.

Limited mobilization costs.

One can both do day and night scanning.

One can view cloud points from any position, not only from the position of the sensor.


D03031/CG9/0M0/001999/ep ARCADIS 66

Risks:

New processes and people needed for data analysis.

Storage and processing of large volumes of point and video data.

Does not work well in rain, snow, fog.

Geometric accuracy decrease when driving in between high rise or tunnels to due to

lack of GPS signal. Can be solved by using Ground Control Points (GCPs).

Evaluation

Costs: Approximately € 100/km. As a rule of thumb every day of data collection will cost

two days of data processing. The cost of purchasing a complete M3DLM vehicle including

all hardware, software and a service contract is € 1,3 M.

Benefits: Mobile laser mapping of objects could replace on-site visual inspections in the near

future. When the imagery is combined, based on the GPS/INS data, a seamless and gapless

recording of the mapped objects is available for in-office inspection. Since images are geo-

referenced, distances can measured on screen and accurate location and size of defects can

be mapped digitally.

Safety: Mobile Laser Mapping is done at traffic speed, hence poses no risk to road users or

inspectors. The laser light used is ‘eye-safe’ and invisible for humans.






Quality: Quality of M3DLM is high, millions of locations are sampled for every metre of

road travelled. The location error of points in the point cloud is in the order of 2,5 cm.

5.5.2 MOBILE VIDEO MAPPING OF OBJECTS

The technological whereabouts and specifications of mobile video mapping are provided in

Section 3.5.2. Mobile 3D Video Mapping is specifically designed for the identification of

objects and to measure features. The mobile video mapping imagery is geo-referenced, so

every pixel has X, Y, Z information with an error of +/- 5 cm or better. The 360º feature of

the system allows the gathering of 3D object information. The relative high position of the

cameras with respect to the safety barriers might allow to view ‘into’ the safety barriers (see

Figure 28).

Fig.: 28. A relative high position of the camera (left picture) might allow to provide information of the

‘inner side’ of the safety barrier.

Mobile 3D Video Mapping seems to be an efficient technique to collect road object

information at driving speed with high location accuracy as well as high resolution imagery.


D03031/CG9/0M0/001999/ep ARCADIS 67

The combination of stereo imaging with GPS/INS positioning makes that road objects

conditions can be directly stored in a geo-database and retrieved in a later phase for instance

for comparison (‘archiving’). It allows more objective inspections of road objects

The 3D video recordings as produced by this system can greatly contribute to Visual Road

Database (Visuele Wegendatabank; VWD) that is under investigation by RWS-DID (ITC

report, 2008). The basic idea of the VWD is to capture all data being necessary to manage

and maintain the Dutch highways by means of optical images or similar appropriate sensing

sources. It is believed that with the VWD the number of field survey can be reduced

significantly and that road maintenance and planning can be improved based on sound

information. The VWD should provide images of various road characteristics like:

hectometer signs (including number identification)

viaducts, bridges, noise protection installations

entries/exits

road surface material (paving)

road markings

vegetation/landscape

crash barriers

traffic signs

road shoulder

illumination installations

Round view and a dense sequence of data capture (< 10 meters) is preferred and can be

provided by the 3D video mapping. The mobile video mapping configuration also gathers

geometric information of the road objects at the same time. With the current video system

pavement deterioration at an early stage like ravelling is difficult to retrieve however. At the

same time one should clearly define the requirements specifications (see also Section 3.3).

The VWD information should be updated at least once a year and immediately after

relevant changes. The VWD information should be accessible from any internal and external

working places. A prerequisite that is also being observed during the Advanced Inspections

Workshop (Reid & Oostrom, 2009).

Since post-processing of imagery by people is still costly and tedious work, a lot would be

gained when objects could be detected automatically in the video imagery. This would

dramatically decrease the number of video stills now need to be processed manually during

‘on-screen’ inspections. Due to the high overlap of the photographic images, objects are

repeatedly present in a variety of frames, which improves the chances of successful

automatic feature extraction (= Computer vision). This task however is still not satisfactorily

solved in computer vision for arbitrary objects in arbitrary situations.

Evaluation

Costs: A brief cost-benefit analysis showed that the total cost (collect video, develop

software, post-processing of images) were higher than when using the classic way of

visually inspecting. Using this method only for maintaining a database would therefore not

be cost-saving.

Benefits: Mobile video mapping of objects could replace on-site visual inspections in the

near future. When the imagery is combined, based on the GPS/INS data, a seamless and

gapless recording of the mapped objects is available for in-office inspection. Since images

are geo-referenced, distances can measured on screen and accurate location and size of

defects can be mapped digitally.

Safety: Mobile video mapping is currently done at speeds of 40 kph, hence a warning for a

slow moving vehicle is needed or a rolling road block is used.



http://en.wikipedia.org/wiki/Computer_vision


D03031/CG9/0M0/001999/ep ARCADIS 68




Quality: Quality of 3D video mapping is high, a total of nine frame cameras are used. The

location error of points in the imagery is in the order of 4-5 cm.

5.6 EVALUATION OF OPPORTUNITY VI TECHNIQUES



Table 22. Categories, inspection requirements and innovations.

Categories Inspections Mobile Laser

Mapping

Mobile Video

Mapping

Safety barriers Gaps + +

Broken, deformed of

cracked components + +

Structure – nuts / bolts /

welds

Corrosion (reinforcement)

Tension

Incorrect height +

Anchorages (i.e. posts) -

movement and ground

condition

+

Ingress of water to post

sockets

Excessive weeds +

Concrete cracks/spalling +

Gantries Welding

Paint damage/rust

Foundation failure

Height + +

Signage Readability/visibility +

White lines Reflectivity + +

Vegetation Length + +



23 below.


D03031/CG9/0M0/001999/ep ARCADIS 69

Table 23. Ranking of techniques for opportunity VI

Automated detection


Technical Availability Overall

Mobile 3D Laser Mapping

1 Extremely rapid

Cheap Never Mostly Available Not used in RWS object inspections

8.6

Mobile 3D Video Mapping

2 Extremely rapid

Cheap Always required

Mostly Available Not used in RWS object inspections

6.8


D03031/CG9/0M0/001999/ep ARCADIS 70

CHAPTER

6 Embedding new technologies

The previous sections provide information about innovative road inspection technologies

and methods. This section discusses who, why, what part and how within the RWS

organization these technologies can or should be embedded. It is clear that RWS - from a

technical and organizational point of view - is looking for new ways to collect, host and

share information about their assets. When discussing the embedding of innovation

inspection technologies it is recommended to (re-)consider the role of the national and

regional institutions as well that of the private market sector. Next to these technical issues

regarding the embedding of new technologies it is also of importance to consider issues that

are related to the adoption of new technologies within the organization and the new ways of

working within the RWS organization. This section addresses organizational issues around

(1) the embedding of new inspection technologies within the RWS organization, (2) the

increasing role of the private market sector in relation to road maintenance and inspection

and (3) change management issues.

It is not the intention to provide final answers in this section regarding the embedding of

new inspection technologies, as this depends on the RWS policy and the direction it wants

to go regarding road asset management and inspections (see Figure 33).

6.1 THE CURRENT SITUATION

The RWS organizations that are responsible for the road conditions are the 10 regional

institutions (see figure 29). Under the regional institutions so called ‘wet’ and ‘dry’ Districts

might reside. Roads obviously fall under the responsibility of the ‘dry’ Districts. Whereas

the Regional institutions and their Districts are responsible for the maintenance of the Dutch

road network, the national institutions provide counsel about policies, standards and

procedures. Furthermore the national institutions are responsible to provide overview of the

road network at a national level and maintain therefore national databases.

Fig. 29 Organization chart of the RWS institutions that are relevant to this study


D03031/CG9/0M0/001999/ep ARCADIS 71

As the regional institutions are responsible for the maintenance of the roads in their region

one would expect therefore that the road inspections are carried out by the regional

institutions (or their subordinate districts) as it would mean that the source of road

condition information is closest to those who are responsible for maintenance. Although

some of the technical inspections are part of the district’s mandate and should be carried out

on a regular basis, practice learns that the regional institutions are carrying these technical

inspections on an ad hoc basis as they are understaffed and absorbed by day-to-day traffic

flow management. A benchmarking report (Kersten and Steenis, in prep.) is expected to

address this point and describes that - except for the ‘Schouw’ (see Figure 32) - the

contribution of the regional institutions to the technical inspections is low. The report

observes that the technical inspections are fragmentized but mainly carried out by the

various national specialist institutions (see 1, Figure 30). Furthermore, the information that

is gathered by the regions during inspections is locally stored and differs from one region to

the other (see 2, Figure 30). Accessibility to the different information sources is crucial to

understand the condition of the road system and includes pavement, structures and other

road objects (see 3, Figure 30). This is of particular importance for the regional institutions.

As this is currently not always the case improved accessibility to the information is

identified as a need for better road asset management (Advanced Inspections Workshop

(Reid and Oostrom, 2009)).

Fig.: 30. This report focuses on the collection of road condition information in relation to decision taking (upper flow diagram). The lower diagram illustrates how currently road

information is being collected and how it affects the decision making within RWS (for more details see benchmarking report (Kersten and Steenis, in prep.).

The information retrieved by the technical inspections is relevant for the national and

regional decision making about budget allocations for road maintenance (see 4, Figure 30).

The above described collection process implies that the actual responsible organization for

road condition management and maintenance (the regions) do not have good access to

proper and up-to date information about the conditions of the road system they are

responsible for. The regional institutions are not able therefore to enter the discussion about

planning and priority settings at the same level as the national organizations. This is not an

ideal situation for the regional managers, nor for the national institutions as information is


D03031/CG9/0M0/001999/ep ARCADIS 72

incomplete and distributed over a variety of databases (see Figure 30) that are not well

connected and stored in a variety of formats.

6.2 THE ROLE OF THE PRIVATE MARKET SECTOR

With the RWS principle ‘the market unless…’, the private market sector will play an

increasing role in the operational asset management activities of RWS including road

inspections (see Figure 31).

================

Fig.: 31. Within the RWS asset management maintenance loop the role of the private market sector will increase. To what extent is still under discussion. In the upper figure the

current RWS situation is given, in the lower diagram the possible future situation for RWS. The ‘Future’ figure also shows the current situation in England.

Source: PIM16

Congress)

This means that the commercial market is not only engaged in the Design & Build phase of

an infrastructure, but that they will increasingly be engaged in Maintenance contracts as

well (DBM contracts).

16 PIM stands for Partnerprogramma Infrastructuur Management and is a 4 years RWS program in order to test

new policy & management procedures.

http://www.projectpim.nl/english/


D03031/CG9/0M0/001999/ep ARCADIS 73

Why is this of importance in the discussion about the embedding of new inspection

technologies? Road inspections are a part of the asset maintenance cycle and RWS is

considering a substantial involvement of the private market sector in the technical

inspections. It is for the technical inspections where the use of innovative new techniques

plays an important role17

. This tendency was reflected in the discussion about ‘Recalibration

of the Market-RWS relation regarding Inspections’ workshop during the PIM congress in

June 2009 (see Figure 32).

Fig.: 32. The private market sector will have a growing role in the technical inspections.

Here an example for the inspection of structures (adapted from ‘Herijking van de verhouding Markt-RWS bij het inspectiewerk’; PIM Congress).

With this new market approach RWS should ask itself the question to what extent the

organization would like to embed new inspection technologies within the organization. An

important question that needs to be answered is what is actually understood by

‘embedding’ in this report? For example are we discussing here the possibility to embed the

entire sensor technology chain; from the purchase and usage of new sensors and platforms

through data hosting, processing and analysis tools, or do we understand by embedding the

enablement to include the final output of these new technologies in the RWS asset

management information and decision structure? Or even one step further; should the

discussion about what type of inspection technologies will be used better be left to the

private market sector as long as the RWS standards regarding safety and road performance

are being met?

As long as the retrieved information is accessible by RWS and can be integrated in its own

information management systems it is the responsibility of the private market sector to host,

maintain and update the information in order to comply with RWS performance contracts.

If in the (near?) future RWS might not have any inspection technology in house at all, one

might wonder why this current investigation should have been conducted on behalf of

RWS-DVS in the first place. There are several answers to that question;

First of all, RWS is still reviewing its current role regarding road inspections. The exact

role of the private sector in this is not yet decided upon and therefore it might be that

RWS decides to keep certain (innovative) inspection technologies in house.

17 In time 24/7 sensor technologies might change this current setup of inspections.


D03031/CG9/0M0/001999/ep ARCADIS 74

Secondly, RWS should remain a knowledgeable institution in order to be able to

appreciate and discriminate between potential new road inspection technologies in

terms of state-of-the-art, performance and compliancy with the RWS (future)

requirements before contracts are being awarded. Therefore RWS needs to understand

what is available in the market on the subject of new road inspection technologies and

understand the pros and cons of a technology, not only in terms of costs but also in

terms of services (to the users), safety and sustainability.

Thirdly, the question arises about the role of RWS regarding innovations? Will the

private market sector put sufficient efforts in innovations if they will become the

principle contractor for road maintenance? Will RWS leave innovations entirely to the

market or should RWS be the inspirer, facilitator and moderator of innovations like, in

this case, road inspection technologies? If the latter is the case, RWS needs to

understand the potential of innovative technologies and be aware of the range of

options.

Fourth, RWS will remain co-operator and the main responsible organization of the

Dutch road system, hence still needs access to the information collected by different

contractors using various inspections technologies. This is in order to understand and

to make a prognosis of short-term (1 to 5 years) and long-term maintenance efforts and

investments and to guarantee safe roads. RWS should therefore be able to understand

how the retrieved information matches the RWS requirements.

Fifth, road inspection technologies are advancing and RWS should be alert of new

advances that might have an impact on their policy and strategy. Future inspection

technologies like motes or ‘Car as Sensor’ for instance provide ‘24/7’ information on the

condition of the entire road network and this will have an effect on inspections,

information requirements, decision making and maintenance procedures.

The role of the public as well as the private market sector regarding maintenance needs to

be reconsidered; who will be responsible of the huge information flow that these

technologies will produce? Who will take care of the privacy issues? How can the user better

be served with real time road (condition) information? At the same time, how can the

responsible road operator take

advantage of this information, as the

same information may serve different

goals (traffic flow and road

condition)? Many questions that are

related to advanced inspection

technologies and need to be solved by

the public sector and the private

market.

If RWS decides to transfer the

technical inspections to the private

market sector it can best relate them

directly with the regional RWS

institutions that are responsible for

road maintenance and shorten the

lines between input (information) and output (performance) road management. With that it

will follow the decentralization approach in the UK where inspection and maintenance is

mainly performed at a regional level by the private market sector and where the UK

Highway Agency has principally a controlling and coordinating tasks (see Figure 33). In

Fig.: 33. The direction that RWS might go regarding road asset management and road

inspections


D03031/CG9/0M0/001999/ep ARCADIS 75

order to establish the right contract it is important thereby that the RWS institutions are able

to understand and communicate with the market its requirements for asset condition

information through inspections.

6.3 MANAGEMENT CHANGES

Embedding innovative technologies within an organization is not just about technology

improvement. It is about capacity development within the organization to adopt new

technologies in the working and decision process and aims at improving the ability of the

entire organization to perform agreed tasks, either singly or in co-operation with others. The

aim of capacity development is to strengthen the organization and includes the involvement

of managers (who will set the requirements to establish what technology is required and

have to decide at the end on the investment in a new technology) as well as the field

workers (who have to work with it). This means that not only technology-oriented

professionals are required in the embedding process, but also staff that can formulate,

design, manage and negotiate within the organization in order to address the following

issues (see also Table 24).

Table 24: Three levels of capacity development

Purpose Focus

Institutional strengthening Strengthening the capacity to develop and negotiate

appropriate mandates and modus operandi as well as

legal and regulatory frameworks

Organizational strengthening Strengthening the management capacity of

organizations by embedding new technologies and

strategic management principles

Human resource development Supply of technical and professional personnel

Institutional strengthening is concerned with the creation of ‘enabling environments’,

i.e. the overall policy, economic, regulatory, and accountability frameworks within

which institutions and individuals operate. Relationships and processes between

institutions, both formal and informal, as well as their mandates, are thereby of

importance.

Organizational strengthening focuses on overall performance and functioning

capabilities, such as developing tools, guidelines and information management systems

for the ability of the organization to adapt to change. It aims to develop its constituent

individuals and groups, as well as its relationship to the outside.

Human resources development aims at the supply of technical and professional

personnel by increasing their knowledge and changing attitudes and behaviours, most

frequently through training and education.

In the scope of the above capacity development requirements, the Development Test Centre

(Dutch: ‘Ontwikkel Test Centrum’,OTC) of RWS may play a role. The OTC is currently under

development, initiated by the PIM program and envisages to add value between the

Technical Conception phase of a new technology and the Business Case (adoption) phase of

the innovation process by taking into account the products, the process and the demand

during an iterative design process (see Figure 34). The OTC can act as a matchmaker

between the market player that offer a new solution and the RWS stakeholder. The


D03031/CG9/0M0/001999/ep ARCADIS 76

Innovation Test Center (ITC18

) of RWS can act as facilitator for such technology innovations.

In all circumstances the new technology should be demand driven. The exact role of OTC

will become clear in the coming year and is expected to have positioned itself before the end

of the PIM program.

Fig.: 34. The potential role of OTC and ITC as matchmaker and facilitator, respectively in

the development and embedding of innovative road inspection technologies. See Figure 30

for a more detailed explanation of the working chain.

18 The ITC of RWS supports the market sector with innovative and promising technologies to test it on

its performance and its costs-benefits. Such validation process is partly financed by RWS.


D03031/CG9/0M0/001999/ep ARCADIS 77

CHAPTER

7 Conclusions

The investigation revealed a myriad of promising technologies to support RWS with the

inspections of the Dutch highway network. It also revealed that several highway agencies

(or related R&D institutions) invest together with the market in the research of new

inspection technologies. Although space borne InSAR and airborne hyperspectral

technology might become useful as hot spot detection technology for deformation and

pavement wearing phenomena over large areas, the most promising technologies are

terrestrial based as they provide the required detail to observe deterioration features. A clear

tendency was found in sensor development to retrieve digital information at traffic speed

for further elaboration of the information in the office. This has advantages in relation to the

visual inspections as the observations will become more objective, is safer for the inspectors

and will lead to less traffic hindrance. If properly archived, the digital information can be

used for retrospective analysis of pavement defects or wearing processes.

Two directions of technology development are observed; those that can detect physical

defects and those that can help predict and prevent physical defects. Most of the

technologies like 3D video, 3D laser, ‘Car as Sensor’ detect the outcome of a wearing process

and observe the physical defects as early as possible. Technologies such as SmartPaint,

‘Smart Dust’ and pavement ‘fingerprinting’ might grow into useful technologies to better

understand the status of wearing before the actual deterioration takes place. Such

‘prevention’ sensor technologies are of major interest to RWS as they provide insight in the

fail processes of pavement, bridges, steel, coatings, etc. and can support RWS in their desire

to improve risk based asset management and safety driven maintenance (outcome of the

Advanced Inspections Workshop; Reid & Oostrom, 2009). To make these sensor

technologies useful for RWS, empirical research should be stimulated that focuses on

composition behaviour in relation to wearing processes. That is the way that the

development of SmartPaint is following and that is the way that other ‘prevention’

technologies should follow as well.

So far, the most feasible sensor technologies will support and decrease the intensity of field

inspections but will not replace them. The current sensor technologies will in particular be

of value to support the annual condition (Dutch: ‘Toestand’) inspections. For the (daily)

Schouw inspections, information is required at a high frequency and at a high intensity that

can not (yet) be delivered by the sensors whereas the maintenance (Dutch: ‘Instandhouding’)

inspections requires more ‘in depth’ information. However, with the ‘Smart’ technology

developments and ‘Car as Sensor’ technologies one might support daily inspection activities

(Dutch: ‘Schouw’) with high frequent information with a high intensity as well as ‘in depth’

wearing processes of road assets for maintenance (Dutch: ‘Instandhouding’).

At this moment there is no sensor available that meets all user requirements. The combined

use of different sensors is still required to cover the needs of the highway inspectors. If, for


D03031/CG9/0M0/001999/ep ARCADIS 78

costs-benefits or other reasons (traffic hindrance, safety), different technologies are being

placed on one platform one should consider if the flexibility of the usage of the sensors is

still guaranteed. Furthermore certain technologies can only operate during day time or

under dry conditions whereas others do have the advantage to operate under all type of

conditions.

Almost all methods indicate that the initial costs for the implementation of new sensor

technologies are higher then the current inspection methods. That may be a reason for RWS,

in particular for structure inspections, to be careful when implementing (in situ) sensors on

a regular basis, as the life span of the structure might not justify the sensor costs. It is sound

that RWS carefully decide whether to implement sensor technology or continue with the

current, mostly visually based, inspection methods. At the same time one should look at the

long-term effects of a technology (in terms of several years or even decades). If a technology

with high initial costs can optimize the maintenance of the road network - like for instance

the timely replacement of particular pavement transects - it might safe tens of millions of

maintenance costs.

Besides the ability to collect pavement data at traffic speed, network wide and monitor

structures ‘24/7’ using dedicated sensors, some further requirements were identified during

the Future Inspection Workshop (Reid and Oostrom, 2009). Better access to information, the

‘right information at the right time’ and (geographic) information systems that can combine

existing data with new data were amongst the wishes of stakeholders. The observed new

technologies all have one thing in common; they provide a huge amount of digital data,

most of which seems to be of not much value since that the majority of Dutch highways are

in good shape. The collection of data is one thing; the retrieval of valuable information from

these data is another. How can we ‘train’ the data to only alarm us when certain threshold

has been reached? How can we handle the enormous data flow that will come with our

desire to monitor the entire Dutch road infrastructure ‘24/7’ at high detail? How can we

automate the processes of deterioration detection in the data that will be collected? Tiny

sensors (motes) are being developed that will signal certain measurements by billions at a

time.

Fig.: 35. About 30 years ago one mainframe served millions of people. We are currently in the 1:1

phase where 1 computer serves one person. In the future several hundreds of tiny dedicated

‘computers’ (motes) will serve one person.


D03031/CG9/0M0/001999/ep ARCADIS 79

‘Car as Sensor’ technology developments eventually will invite the engagement of the

public to help RWS to better manage the highways for the good of everyone. This means

that road condition information will flow from and to hundreds of thousands of cars in real

time. These developments are very promising and without doubt it will become the future

for road inspections. In this era millions of digital sensors will serve the public (see also

Figure 35). It is the challenge for the highway agencies to find ways of how to pick up

promising technologies, how it can be tested to understand its potential to introduce new

inspection technologies. It is inevitable that a close collaboration with the market is essential.


D03031/CG9/0M0/001999/ep ARCADIS 80

CHAPTER

8 Recommendations

8.1 INTRODUCTION

In this section recommendations are given about some of the most promising technologies

that are being observed. From the Review & Evaluation Meeting it was learned that RWS is

looking for promising and new techniques that will become available for testing in the next

years. At present RWS is already testing some new ideas (e.g. laser line scanning for

ravelling detection). Also some recommendations are given about the embedding process of

new technologies with the RWS organization.

8.2 MOST PROMISING INSPECTION TECHNOLOGIES

Below a description is given of new technologies that seem most promising for future

inspections.

8.2.1 ‘CAR AS SENSOR’

This innovative concept is among one of the most promising technologies to support the

Dutch highway agency with the detection of physical pavement defects. It is believed that

this technology is not far away (in terms of some years) from becoming an applicable

technology to support RWS in the detection of physical pavement defects at a national level.

If sufficient knowledge of composition behaviour of road elements (asphalt, concrete, steel)

in relation to traffic dynamics is attained, ‘Car as Sensor’ technology can also indirectly

contribute to the prevention of physical defects by delivering location specific information

nation wide about loading, traffic intensity, etc. This type of information can then be used

in models to calculate the actual conditions of the pavement and structures over time which

is another type of identified information need.

R&D set up: RWS and TNO are about to define a research program during the 22nd

of

September kick-off workshop. Ravelling is one of the most important pavement

deterioration features of ZOAB. The Dutch highways constitute for 80-90% of ZOAB and

there is no technology yet available to properly detect ravelling in an early stage. Therefore,

it is recommended to focus with the planned research on the detection of ravelling (location

and intensity). The most promising sensors seem to be those present on the shock

absorbers, the tyre based sensors and in addition acoustics sensors. The potential of ‘Car as

Sensor’ Technology is huge as well as its challenge to channel the huge amount of data it

will eventually produce. Once certain in-car sensors have proved to be successful with the

TNO test bed (see also Section 3.5.1), it is recommended therefore to implement this

technology using the existing RWS fleet and store the data with an in-car device to avoid

communication transfer issues. If deemed necessary - for instance because more dense and


D03031/CG9/0M0/001999/ep ARCADIS 81

frequent information is required - one can extend the technology to the commercial

transport sector (explaining the win-win situation), then to the pubic sector. It is also

recommended to closely follow the ‘Car as Sensor’ technology developments in the

automotive industry and how it is being used in the Dynamic Traffic Management (DVM)

sector as they are the origin of new developments.

Compliancy with need: The further investigation of ‘Car as Sensor’ Technology is compliant

with the findings of the Advanced Inspections Workshop (Reid and Oostrom, 2009) where

the need of public participation in the determination of road quality levels is recognized as

important. The proposed research subject – detection of ravelling - is also compliant with

the identified information needs. In addition it might contribute to other identified

information needs as well like the detection of bumps in the road surface and to indirectly

check the steel joints connections19 between the road and the bridge.

8.2.2 ‘SMART DUST’

‘Smart Dust’ technology (see also Section 3.7.1) is a very promising ‘prevention’ technology

but is still in an early exploration phase. The real value of ‘Smart Dust’ opposed to the

current in situ sensors is the use of a large amount of cheap and tiny sensors that can be

embedded in the material and remotely signal relevant changes over time within pavement,

structures or other road objects which can not be detected visually. SmartPaint technology

can be considered as a specific type of ‘Smart Dust’ technology to monitor the gradual

change of coatings. It is recommended that the development of SmartPaint technology is

closely followed by RWS as it might be an early adaptor of ‘Smart Dust’ technology for

coating inspections.

Compliancy with need: ‘Smart Dust’ technology complies with several of the information

needs that are being mentioned during the Advanced Inspections Workshop (Reid and

Oostrom, 2009). If fully developed the ‘Smart Dust’ technology can provide the right

information at the right time, it can provide location based real time information, the

technology provides high density and high intensity surveillance capacity of the road

conditions.

8.2.3 TAGGING ASSETS IN RELATION TO 3D INSPECTION TECHNOLOGIES

Mobile 3D Laser and Mobile 3D Video Mapping are very interesting ways to identify

components, get insight in its origin and history and to follow status information in the

field. Although it is actually not an inspection technology it definitely supports the

inspections and asset managers with their work. The tagging of assets and the possibility to

obtain object information by using barcodes or through RFID technology is an existing

challenge. The challenge is the implementation of the ‘tagging’ technology in the entire

workflow; from 3D CAD design to 3D Asset Management. Some experience with

engineering bureaus like Ballast Nedam and ARCADIS with 3D CAD design learns that

these types of new technologies effects the working procedures. By using location based

19 A flexible construction with steel joints is connecting a bridge with the road. If the rubber parts within

the joints are breaking down, the steel components become exposed and are ‘hit’ directly by the traffic.

This may damage the connection further, increases the noise and is a hindrance for the traffic (tyres).

Using noise sensor or sensors on the absorbers might be a useful tool to measure the status of steel

joints.


D03031/CG9/0M0/001999/ep ARCADIS 82

service technology20

like RFID and PDA technology information from and to the field

inspectors. One should be aware that the tagging solution should be robust and that the

‘tags’ of the components should be easily accessible by the inspectors.

R&D set up: To take optimum advantage of asset tagging, the entire chain should be

involved in the pilots; 3D CAD designers, BIM experts as well as RWS and Market asset

managers, the manufacturers and the RFID technology sector. It is understood that ample

experience of component labelling technology is present in the aircraft industry so it is

recommended to consults this sector to better understand the pros and cons of the

technology. 2009 and 2010 is a perfect timing to test these approaches as large structural

works will be build. The more as 3D CAD and GIS systems are under development.

Compliancy with need: The many interviews and discussions with inspectors reveal that it

is difficult to retrieve historical information of the road components that need to be

inspected. One does not know if or where the information is available. If available the

information is often incomplete or exists in a format that is hard to work with. During the

Advanced Inspections Workshop (Reid and Oostrom, 2009) the labelling of components is

identified as one of the information needs. Another identified need during that workshop

was the need of access to information at any place and any time. Location based technology

and labelling the components makes this possible. Tagging also allows the combination of

existing information like for instance present in databases like DISK and KernGIS, with

newly retrieved component information that is retrieved during the inspections.

8.2.4 MOBILE 3D VIDEO AND 3D LASER MAPPING

These are two of the most feasible type of pre-operational technologies that are currently

being introduced on the Dutch highways to support inspections. First experiences learn that

the mobile 3D laser technology is well received by measuring existing structures

(Coentunnel, Ketelplein). Whether this technology can also be used for the detection of

pavement wearing features like longitudinal and transversal profiling would be of interest

as well to what extent this technology can be applied for object monitoring like gantries and

portals. More should become known about the possibilities of mobile 3D video mapping as

well, in particular for structures. Structure inspectors find the 3D video technology

appealing but that it currently does not sufficiently replace the visual inspections of

structures as automatic feature detection is still underdeveloped. As both technologies

provide geo-referenced information and imagery capacity (laser with video) it is believed

that both Mobile 3D Mapping technologies are useful instruments to populate the envisaged

Visual Road Database (Visuele Wegendatabank; VWD) at a national level.

R&D set up: Two type of investigations are being proposed to investigate the usefulness of

3D laser & 3D video mapping

1. Coupling of the 3D laser and video data with the envisaged VWD or with the existing

databases KernGIS to strengthen the visualization of the road network and support

inspections, select thereby a road system in order to

2. study the use of the 3D imageries to identify physical wearing features of pavement

(cracks, rutting, profiles, and ravelling), structures (deformations), and other road

objects like gantries, barriers, signage, white lines and vegetation).

Compliancy with needs: There is a clear need of inspection technologies that can contribute

to a better understanding about the status of road objects like gantries, safety barriers, etc.

20 A location-based service (LBS) is an information service, accessible with mobile devices through the

mobile network and utilizing the ability to make use of the geographical position of the mobile device.


D03031/CG9/0M0/001999/ep ARCADIS 83

RWS-DID is investigating the set up of a Visual Road Database (VWD). The basic idea of the

VWD is to capture all data being necessary to manage and maintain the Dutch highways by

means of optical images or similar appropriate sensing sources. Mobile 3D Video and Laser

Mapping techniques allow the combination of existing information that is present in the

various national and regional databases with new information. The speed of information

gathering allows a higher inspection frequency including those of secondary roads. Both last

requirements were identified during the Advanced Inspections Workshop (Reid and

Oostrom, 2009).

8.2.5 RADIOMETRIC FINGERPRINTING

This technology (see also Section 3.7.2) might be an interesting new mobile inspection

technology to provide information about the pavement composition and to support the

prevention of physical defects like ravelling, potholes, cracks, etc. Ultimately the

information about the pavement composition status should provide asset managers the

optimum timing of pavement replacement. However, more investigation is needed to

understand the relation between the measured isotopes, their representation of pavement

composition and the relation between changes in pavement composition and the pavement

wearing process.

R&D set up: An expert meeting should be organized between pavement experts and the

fingerprinting sensor experts to better understand the potential of the fingerprinting

technology in relation to asphalt wearing processes. If deemed useful an investigation

should be formulated where pavement and sensor specialists collaborate. Most useful seems

to be an empirical study that relates measurable changes in the pavement composition with

wearing processes. Can the fingerprinting technology already be helpful in the quality

assessment of asphalt during paving? Can the SmartPaint approach as function as a

blueprint for a study of asphalt behaviour?

Compliancy with needs: Fingerprinting technology can contribute to a better understanding

in an objective and quantitative way the conditions changes of asphalt. With a better

understanding of the wearing processes in relation to pavement composition the technology

can provide the right information at the right time to replace (or not) pavement. The

requirements have been identified during the Advanced Inspections Workshop (Reid and

Oostrom, 2009). It can have a substantial impact on the pavement quality of the highways

and improve cost-benefits and sustainability of the road surface.

8.2.6 INSAR TECHNOLOGY

The space borne technique might be of interest for nationwide hotspot detection of structure

deformation form space. The technology is available and in use for other domains like urban

subsidence. The TU Delft and NITG-TNO have validated the InSAR technology for the

European Space Agency (ESA) for some Dutch urban areas in Alkmaar & Amsterdam

(Hanssen et al., 2008). Problems observed with this technology is that it is not always clear

what point of a structure is functioning as a scatterer so it is also unclear if that point indeed

is representative for measuring deformation. Another problem might be the fact that a

structure does not reveal sufficient scatterers because of its nature or its position with

respect to the satellite. Both problems can be overcome by placing so called corner reflectors

but will increase the costs.

R&D set up: If RWS is interested to understand more about the potential of this technology

it is recommended to set up an InSAR expert meeting with people of TNO & TU Delft and


D03031/CG9/0M0/001999/ep ARCADIS 84

some senior road experts (pavement, structure & objects) to oversee the full scope of InSAR

technology for nationwide deformation measuring for the Dutch road network. As TU Delft

does posses of a considerable archive with historical SAR images of the Netherlands some

pilots might give answer on some of the following critical notes; is the InSAR technology

(including new sensors like Radarsat-2 and ALOS-PALSAR) a mature technology to

measure with sufficient accuracy (meaning sufficient scatterers per object with sufficient

accuracy) deformation in the Dutch road system.

Compliancy with needs: InSAR can collect nationwide information about deformation in a

very short timeframe (monthly) and can be considered as a potential technology for hot spot

deformation detection. It is therefore compliant with the information need to collect lots of

data without much detail to alert managers of pressing problems. The retrieved geo-

referenced information is easily accessible and can be combined with the existing databases

of RWS.

Sustainability: A special note to RWS is to pay attention to the sustainability of an offered

technology. Per January 2010 the sustainable purchase of goods and services (Dutch:

‘duurzaam inkopen’) by the Dutch government is mandatory and will play an important role

in deciding what product to buy. It is recommended to RWS to consider a paragraph in

their tenders and formulate criteria regarding the sustainability of an offered technology.

Mobile mapping technologies for instance offer sustainability as information is collected at

traffic speed, causing less traffic hindrance and therefore reduces CO2 emissions. Inspection

technologies might be offered that will optimize the timely replacement of pavement which

will not only reduce costs and increase safety but also avoid the unnecessary replacement of

pavement which is still in a good condition. It is also of importance for RWS to consider

what possible increase of costs is acceptable to apply sustainable solutions.

8.3 EMBEDDING NEW TECHNOLOGIES AT RWS

Recommendation 1: The discussion about embedding innovative inspection technologies

should not be about how to integrate the technological acquisition and processing line

within RWS. The focus of RWS should be on how to remain a knowledgeable institution

about the market position of the offered solutions and inspirer of the development of new

technologies.

Recommendation 2: As the regional RWS institutions are responsible for road maintenance

they should be responsible to carry out the inspections. It is recommended that they are

being resourced with the proper means21

to carry out the road inspections. The national

institutions like DI, DVS & DID should safeguard standardization issues regarding

information retrieval and be responsible to connect the regional databases to understand the

condition of the road system at a national level which is required for the coordination and

arbitration regarding prioritization and budget allocations. The national institutions provide

counsel to the regional institutions regarding new inspection technologies and are the

inspirer and facilitator of innovations.

Recommendation 3: With the RWS principle ‘the market unless…’, the private market sector

will become the main actor in the maintenance of the Dutch highway infrastructure.

Therefore, and as long as the RWS technical and functional standards are being met, RWS

21 Resourced’ can mean that the regions decide to let the private market perform the inspections


D03031/CG9/0M0/001999/ep ARCADIS 85

should leave it up to the private market sector what type of inspection technologies will be

used to retrieve the information.

Recommendation 4:: The Development Test Centre (Ontwikkel Test Centrum; OTC) should

be involved in the development of innovative road inspection technologies by focusing on

the user demand compliancy of the offered product and on the embedding of these

technologies and/or its derived products into the RWS national or regional organizations.

The ITC should be the facilitator.

Recommendation 5: Based on the results of the investigation and the Review & Evaluation

Meeting (see also Annex 6) it is recommended to elaborate the role that RWS will take

regarding innovations and embedding of innovative inspection technologies within its

organization. With the increased involvement of the market regarding road asset

management, it is uncertain what position RWS will take regarding the stimulation of

innovation. Notwithstanding the desire to stay involved in new technology developments it

is unclear if RWS will take the position of stimulator, facilitator or consumer of innovative

technologies and how the embedding process within the RWS organization will take place.

Recommendation 6: Another recommendation from the Review & Evaluation Meeting (see

also Annex 6) is to further investigate possible business cases of identified inspection

technologies. It is of importance to better understand to what extent, sometimes costly,

innovative inspection technologies can contribute to inspections in relation to life cycle costs

and the expected technical and functional life span of road assets.


D03031/CG9/0M0/001999/ep ARCADIS 86

References

Bakker J.D., J. Volwerk, J. & Verlaan J. (2006). Maintenance management from an economical

perspective, IABMAS.

Bakker, J.D. (2004). Monitoring of ASR expansion and moisture in concrete. 12th

International Conferences on Alkali-Aggregate Reaction in Concrete (ICAAR).

Bakker, J.D. (2008). Control of ASR related risks in The Netherlands. CUR recommendation

89.

Eijbersen, M.J., (2004). Detection fraying through texture laser measurements.

http://www.onderzoekinformatie.nl/en/oi/nod/onderzoek/OND1284586/

Gerke, M. and Prof. Vosselman, M.G., (2008). Methoden en Technieken Visuele

Wegendatabank, ITC report

Hanssen, R.F., et al.. Validation of existing processing chains in TerraFirma stage 2. GMES

Terrafirma project. ESA-ESRIN contract no. 19366/05/I-E

Herold, M., Roberts, D.A., Noronha, V., Smadi, O. (2008). Imaging spectrometry and asphalt

road surveys. ScienceDirect, Tranportation Research Part C 16, pp 153-166.

Herold, M., Roberts, D.A. (2005). Mapping asphalt road conditins with hyperspectral remote

sensing. 5th International Symposium Remote Sensing of Urban Areas (URS 2005), Tempe,

AZ, USA, March 14-16 2005

INTRO (2008a). Intelligent Roads, Final Summary. Intelligent Roads, EU Project no. ST4-CT-

2005-012344

INTRO (2008b). Demonstration of Methods for the Measurement of Condition using Probe

Vehicles. Intelligent Roads, EU Project no. ST4-CT-2005-012344

Kersten, H.,Steenis, M (in prep.). Benchmark Highway Agency versus RWS processes and

systems. RWS PIM report

McRobbie S., Lodge R., Wright A. (2007). Automated inspection of highway structures –

Stage 2. TRL study. Published Project Report 255.

McRobbie S. (2008). Automated inspection of highway structures – Stage 3. TRL study.

Published Project Report 338.

Noortwijk, J.M., Klatter, H.E. (2004). The use of lifetime distributions in bridge maintenance

and replacement modelling. Computers & Structures, 82(13-14):1091-1099.

Pascucci, S., Bassani, C., Palombo, A., Poscolieri, M., Cavalli, R. (2008). Road Asphalt

Pavements Analyzed by Airborne Thermal Remote Sensing: Preliminary Results of the

Venice Highway. Sensors 2008, 8, 1278-1296.

Reid, R., Oostrom, M. (2009). Advanced Inspections Workshop; It stand or falls with good

information’. RWS Internal Workshop report

Sukum Kim et al., (2007). Health Monitoring of Civil Infrastructures Using Wireless Sensor

Networks. 6th International Conference on Information Processing in Sensor Networks

(IPSN’07)

Tao, C.V. (2006). 3D Data Acquisition and Object Reconstruction for AEC/CAD. In:

Zlatanova, S. and Prosperi, D. (eds.): Large-scale 3D data integration – Challenges and

Opportunities. Taylor&Francis, London, pp. 39-56.


D03031/CG9/0M0/001999/ep ARCADIS 87

ANNEX 1 List of People Interviewed

Table 1-1. List of people interviewed for the Advanced Inspections Project

Inspection Contact Organisation Telephone Email

General Erik Vos RWS +31 88 79 82 243 [email protected]

Pavement Wim van Ooijen RWS +31-88-7982222 [email protected]

Structures,

Gantries

Fons Bots RWS +31(0)88-7973510 [email protected].

nl

General Govert Sweere RWS - [email protected]

General Jan van de Zwan RWS +31(0)15 251 83 91 [email protected]

General Max Horsten RWS (District

Zuid Holland)

+31(0)10 402 62 01 [email protected]

General Erik Baak RWS (Dictrict

Zuid Holland)

+31(0)10 402 62 00 [email protected]

General Rajesh Sukhraj RWS +31(0)88-7982298 [email protected]

General Bert de Wit RWS +31(0)15 251 82 06 [email protected]

General Martijn Koster RWS +31(0)15-2518517 [email protected]

General Leo Reijnen RWS - [email protected]

Structures Jaap Bakker RWS - [email protected]

DVM Hans Verschoor RWS - [email protected]

General Bert Elbersen RWS +31(0)610133714 [email protected]

Structures Joost Wäckerlin RWS +31(015) 269 13 59 [email protected]

Sensor Arie Penning RWS - [email protected]

DRIP Henk van Hulst RWS - [email protected]

Structures Herbert Floor RWS - [email protected]

PIM Henk Kersten RWS - [email protected]

Pavements Han Limburg Medusa

Explorations

+31(0)505770280 [email protected]

Pavements Peter Paul

Schachmann

TNO +31(0)887982239 [email protected]

Pavements Martin Herold ESA GOFC-

GOLD Land

Cover Project

Office

- [email protected]

Pavements Simone Pascucci National

Research

Council,

Institute of

Methodologies

for

Environmental

Analysis

- [email protected]

mailto:[email protected]















D03031/CG9/0M0/001999/ep ARCADIS 88

Inspection Contact Organisation Telephone Email

Pavement Seirgei Miller TU Twente +31(0)534894601 [email protected]

Structures Michiel de Jong,

Wouter Zomer

Koenders

Instruments,

IJkdijk

+31(0)365486180,

+31(0)613973507

michiel.dejong@koenders-

instruments.com;

[email protected]

Pavements Monica Starnes TRB +1 202 334 1894 [email protected]

Pavements Prof. Andre

Molenaar

TU Delft - [email protected];

General Cheryl Richter US DOT - [email protected]

GPS - Galileo Oliver

Juckenhoeffel

Astrium /

EADS

+49 8960720509 [email protected]

ads.net

General Roland

Spielhofer

Austrian

Institute of

Technology

+43 505506284 [email protected]

General Marek Truu Technical center

of Estonian

roads

+372 677 1500

[email protected]

‘Car as

Sensor’

Michal Even-

Chen, Shimon

Nesichi

Iroads, Israel +972 37355876 [email protected];

[email protected]

General Martin Holt

Martin Senior

Infoterra UK +44 116 273 2331 martin.holt@infoterra-

global.com;

martin.senior@infoterra-

global.com

Geo-detect Marcel

Bastiaansen

ARCADIS NL +31 (0)627 060848 [email protected]

l

Pavements Jaap van de

Gevel


DVM Vincent van der

Heijden

ARCADIS NL - vincent.vanderheijden@arcadi

s.nl

Mobile Laser

and Video

Mapping

Joost Swarts ARCADIS NL +31 (0)627 062131 [email protected]

RWS Geo-

informatie

Klaas Jan Sant ARCADIS NL +31 (0)627 061399 [email protected]

Pavement

Structures

Léon Haarman ARCADIS NL +31 (0)627 060378 [email protected]

Geo-detect Marcel

Bastiaansen


l








http://portaal.arcadis.nl/SearchCenter/Pages/PeopleResults.aspx?k=WorkPhone:%22%2B31%20%280%2962%207060%20848%22




https://email.arcadis.nl/OWA/redir.aspx?C=c6d30cc912d747568a249f2b4f8a94d1&URL=mailto%3aklaasjan.sant%40arcadis.nl


http://portaal.arcadis.nl/SearchCenter/Pages/PeopleResults.aspx?k=WorkPhone:%22%2B31%20%280%2962%207060%20848%22




D03031/CG9/0M0/001999/ep ARCADIS 89

ANNEX 2 Aran

The Dutch ARAN-3 vehicle is equipped with the following systems:

POS/LV 420 (INS)

GPS

Longitudinal profiler

Laser XVP (for the road surface only)

Pavement video (1 mm resolution)

Smart Texture

Panoramic ROW video Pictures (full colour, resolution 1920 x 1080, objective angle 91

degrees, shutter time 1ms). Refer to Figure 2-1 below for an example image.

Visidata (software tool for comprehensive viewing of data)

Wisecrax (software for crack detection)

Surveyor 2.0 (software for asset inventory)

Fig.: 2-1. Example image from the ROW camera provided by Ir. W.H. van Ooijen

(source: Geo-Info 2008-6)


D03031/CG9/0M0/001999/ep ARCADIS 90

ANNEX 3 ‘Car as Sensor’

In this report the term ‘Car as Sensor’ is frequently referred to. This appendix gives a brief

overview what is actually meant by ‘Car as Sensor’. When we look at ‘Car as Sensor’

technologies it is particularly focusing on traffic flow management and in car safety issues.

As such it can be considered as part of an Intelligent Transportation Systems (ITS). ITS (see

also Wikipedia: ITS) refers to efforts to add information and communications technology to

transport infrastructure and vehicles in an effort to manage factors that typically are at odds

with each other, such as vehicles, loads, and routes to improve safety and reduce vehicle

wear, transportation times, and fuel consumption. Traffic flow conditions and safety issues,

as mentioned under ITS, are strongly linked to ad hoc (snow, ice, and water) and long term

(ravelling, rutting) surface and road conditions. As such it is a logical step to consider ‘24/7’

road condition monitoring as an important element of ITS to optimize the management of

the roads by improving the asset knowledge, contribute to safety on the road, increase of

maintenance efficiency, and reduce traffic hindrance. As there is a common interest in using

in car systems for ITS purposes it makes sense to consider the potential of the ‘Car as Sensor’

for road condition monitoring as well. Moreover, the automotive industry is already

introducing electronic equipments in cars to collect data of road conditions for safety issues

(see Table 3-1)

Table 3-1. Current and future in-car systems and sensors as being of potential for road condition

inspections (adapted from the INTRO report (INTRO, 2008b)

*HGV=Heavy Goods Vehicle

http://en.wikipedia.org/wiki/Intelligent_transportation_system


D03031/CG9/0M0/001999/ep ARCADIS 91

Communication Technology

‘Car as Sensor’ is not only about the sensor technology. Of equivalent importance and a big

challenge is how the data will be collected, analyzed and disseminated. The majority of

modern vehicles share data from such sensors across different parts of the vehicle using the

Control Area Network (CAN) but the information is not logged. There is potential to either

store the data for offline assessment of pavement condition in data loggers or to transfer the

data from the CAN for real-time assessment; Intelligent Co-operative Systems. Intelligent

Co-operative Systems will build and expand on the functionality of the autonomous and

stand-alone in-vehicle and infrastructure-based systems. The systems that are based on

vehicle-to-vehicle (V2V) and Vehicle to Infrastructure (V2I) communications hold the

promise of great improvements both in the efficiency of the transport systems and in the

safety of all road users. In both flows the car is the focal point for information retrieval and

reception. To allow the communication and transfer of information between vehicles and

the road operator, information models should be developed to handle the massive data

flow. The European CVIS project (Cooperative Vehicle Infrastructure Systems) addresses

issues that are related to getting data back to asset managers from vehicles.

Organizational impact

The huge amount of information that will be collected, manipulated and disseminated will

become a big challenge for any highway agency and their contractors. They should face this

challenge together with ICT partners, the navigation industry and the automotive industry.

Once ‘Car as Sensor’ technology has reached an operational stage it will affect the current

inspection strategy and have an impact on the organizational structure of RWS. Privacy

issues will play a role as well as the observation will only become valuable when it is linked

with a location. Although the highway agency like RWS is not interested in who is sending

or receiving the information it should think about how to convince the public to share the

information with the road manager.

http://www.cvisproject.org/


D03031/CG9/0M0/001999/ep ARCADIS 92

ANNEX 4 Scoring of Techniques

Each technique has been scored according to a series of indicators, which relate to the

technical abilities of the technique to fulfil the RWS inspection requirements as outlined in

Sections 3.2, 4.2 and 5.2, and data acquisition scores. The scoring has been carried out

according to the details set out in the tables below (Table 4-1, 4-2, 4-3, 4-4 and 4-5). All of the

techniques investigated have then been ranked, to give an indication of the most

appropriate techniques for each opportunity) see also Table 4-6).

Table 4-1 Technical scoring for each technique relative to inspection requirements

Score Description

3 Inspection requirements fully covered

2 Inspection requirements mostly covered

1 Inspection requirements partially covered

0 Inspection requirements not covered

5.1.2 Data acquisition scoring

Each technique has been scored for three indicators that relate to the acquisition of the data:

speed of data acquisition in the field (considered in terms of length of RWS road)

cost (excluding traffic management)

requirement for traffic management during data acquisition using the technique

The scores allocated have been determined according to the frameworks set out in Table 4-

2a for pavements and 4.2b for structures (speed), Table 4-3a for pavements and 4-3b for

structures (cost), Table 4-4 (traffic management requirements). It should be noted that these

scores, particularly speed and cost, are order of magnitude figures only, and are intended to

give an indication of the relative costs and speeds of techniques when compared to each

other. The speed and cost scores for each technique are likely to vary according to a wide

range of variables, such as different suppliers, the environment in which they are used,

weather conditions during survey, etc.

Table 4-2a Data acquisition scoring – speed of acquisition in the field (pavements)

Score Description Indicative Range

5 Extremely rapid: 100s of kilometres per day or more

4 Rapid: 10s of kilometres per day

3 Moderate: 1000s of metres per day

2 Slow: 100s of metres per day

1 Very slow: 10s of metres per day or less

Table 4-2b Data acquisition scoring – speed of acquisition in the field (structures)


5 Extremely rapid: 1000 structures per day or more

4 Rapid: 100 structures per day

3 Moderate: 10 structure per day

2 Slow: 1 structure per day

1 Very slow: 0.1 structure per day or less


D03031/CG9/0M0/001999/ep ARCADIS 93

Table 4-3a Data acquisition scoring – cost of acquisition (excluding traffic management

costs) for network inspections (pavement and objects)


5 Very cheap €20 per kilometre or less

4 Cheap €200 per kilometre or less

3 Moderate €2,000 per kilometre or less

2 Expensive €20,000 per kilometre or less

1 Very expensive €20,000 per kilometre or more

Table 4-3b Data acquisition scoring – cost of acquisition (excluding traffic management

costs) for structure inspections


5 Very cheap €20 per structure

4 Cheap €200 per structure

3 Moderate €2,000 per structure

2 Expensive €20,000 per structure

1 Very expensive €20,000 per structure

Table 4-4 Data acquisition scoring – requirement for traffic management

Score Description Comment

3 Never required Traffic management is never required (generally

applies to aerial, remote captured data acquisition

techniques)

2 May be required Traffic management may be required, depending on

the local conditions and location of the application of

the technique

1 Always required Traffic management will always be required.

5.1.3 Availability and status scoring

Each technique has been scored in terms of the availability of the technique (commercially,

and whether available in the Netherlands or UK) and the status of the technique (whether it

has been used in the RWS or HA inspections or not). The scores have been assessed in terms of

the descriptions outlined in Table 4-5.

Table 4-5 Availability and status scoring

Score Description

4 Commercially available in the Netherlands

3 Commercially available in the Netherlands/not or little used in RWS environment

2 Commercially available outside the Netherlands

1 Not commercially available/in research, development or prototype

0 Not available / only in the minds of visionaries

5.1.4 Overall ranking of techniques

Using the scores assigned to each technique, a ranking has been produced to identify the

most appropriate technique for achieving the required inspection results for each identified

opportunity. It should be noted that availability has not been included in the score; this is to


D03031/CG9/0M0/001999/ep ARCADIS 94

ensure that promising techniques which have yet to reach market can be considered on their

merits too. However, the summary tables include the availability for comparison purposes.

The overall score for each technique, per knowledge step, is allocated according to the

following formula: technical score × ((speed score × 0.5) + (cost score × 0.3) + (traffic

management score × 0.2)). The highest possible overall score is 13.8. The scoring of each

technique per opportunity is given in Table 4-6 below.

Table 4-6: Ranking of techniques per opportunity.

Pavements

Opportunity I Technique Ranking Technical Speed Cost TM Overall

‘Car as Sensor’ 1 2 5 4 3 8.6

Mobile Video mapping 2 2 5 4 2 8.2

UAV 3 1 5 3 3 4

Space Borne 5 0 5 5 3 0

Vehicle based GPR 3 1 5 3 3 4

Opportunity II

Technique Ranking Technical Speed Cost TM Overall

Smart dust 1 2 5 3 3 8

Medusa fingerprint 2 1 5 4 3 4.3

Hyperspectral 3 1 5 4 3 4.3

ASPARI, TU Twente 4 1 2 3 3 2.5

Structures Opportunity III Technique Ranking Technical Speed Cost TM Overall

RFID tagging 1 1 4 2 3 3.2

Cable scan 2 1 3 3 3 3

Digital photography 3 1 2 3 3 2.5

Opportunity IV


In-situ sensors 1 2 5 1 3 6.8

Opportunity V


Smart paint 4 1 3 3 5 3.4

Fiber optics – FBG’s 2 1 4 2 5 3.6

Deformation sensor 2 1 4 2 5 3.6

InSAR 1 1 3 4 5 3.7

Objects Opportunity VI Technique Ranking Technical Speed Cost TM Overall

Mobile Laser Mapping 1 2 5 4 3 8.6

Mobile Video Mapping 2 2 4 4 1 6.8


D03031/CG9/0M0/001999/ep ARCADIS 95

ANNEX 5 Evaluations

To provide a rapid overview of the status of innovative technologies in this project,

summary evaluations are provided in this annex. The stage of development of the technique

is indicated by a colour bar (see Figure 5-1 below).

Fig.: 5-1. Example colour bar used to indicate the operational status of each technique

5.1 ‘Car as Sensor’

General

Feedback from users TNO is currently testing (Guido Sluijsmans)

Advantages ‘24/7’, network wide coverage, statistical

testing possible

Multi-purpose: longitudinal evenness,

ravelling?, rutting, loose objects

Huge number makes that some failing

sensors do not affect the monitoring

Disadvantages Dependent on car makers, unless only RWS

fleet vehicles

Large data volumes

Quality of data lower than dedicated sensors,

good enough for pin pointing trouble spots

Technical scoring

Inspection requirements mostly covered

Data acquisition scoring

Speed of acquisition Extremely rapid

Cost of acquisition Unknown

Traffic management Not required

Availability and status scoring

Availability and status Not available in the Netherlands. Not yet in R&D

stage.

Recommendations/comments

Recommended for further testing Yes

Recommended for use Yes, start with a pilot using RWS fleet of cars

Other comments Take into account huge dataflow


D03031/CG9/0M0/001999/ep ARCADIS 96

5.2 Mobile 3D Video Mapping

General

Feedback from users Joost Swarts (ARCADIS NL) is testing the system

Advantages Rapid surveying

Accurate, 3D geo-referenced imagery

Objects can be measured on screen

Disadvantages Not suitable in rain or mist

Obstruction by other vehicles

Technical scoring



Speed of acquisition Extremely rapid, post-processing: 2 days/day

Cost of acquisition Cheap: € 100 per kilometre



Availability and status Commercially available in the Netherlands, R&D


Recommended for further testing Yes, in particular to test automated feature

extraction

Recommended for use Yes, as part of cross asset survey, rather than for

pavements alone

Other comments Traffic management issue needs to be resolved


D03031/CG9/0M0/001999/ep ARCADIS 97

5.4 UAV

General

Feedback from users No feedback was received from user

Advantages Allows for quick and detailed imaging of

road surface conditions

Disadvantages Low location (X,Y,Z) accuracy

Safety, what if UAV crashed on or near a

highway? Model airplane laws apply

Operational time (30 mins - 2 hours max)

Technical scoring

Inspection requirements partially covered



Cost of acquisition Cheap



Availability and status Commercially available. In R&D stage.


Recommended for further testing No

Recommended for use No

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 98

5.5 Remote Sensing

General

Feedback from users No feedback was received from user

Advantages Large area overview

Disadvantages Too low image resolution for inspection

requirements

Technical scoring

Inspection requirements not covered






Availability and status Imagery is readily available, not used in RWS




Other comments


D03031/CG9/0M0/001999/ep ARCADIS 99

5.6 Ground Penetrating Radar (GPR)

General

Feedback from users Yes, Leon Haarman (ARCADIS NL) is using GPR

successfully in rail road bed inspections. While

Breijn, the R&D department of Heijmans

pavements construction company provided

feedback on the successful use in a highway

scanning project (A12 at Zoetermeer).

Advantages Can indicate problem areas, voids and

subsurface irregularities

Can be done at traffic speed

Disadvantages Cores are still needed to check the exact

nature of the problem

Not possible to distinguish two asphalt

layers, since material properties are too

similar

Technical scoring



Speed of acquisition Rapid


Traffic management Not required when air-coupled high-speed GPR is

used


Availability and status Available in the Netherlands, R&D




Other comments


D03031/CG9/0M0/001999/ep ARCADIS 100

5.7 ‘Smart Dust’

General

Feedback from users No feedback received

Advantages Vehicle independent

Multi purpose: measure traffic dynamics,

loads, temperatures

Huge number makes that some failing

sensors do not affect the monitoring

Disadvantages Yet unknown when these sensors should be

put in the asphalt mix, so as not to be

destroyed by high paving temperature.

How to supply power once the sensors are in

the asphalt layers?

Technical scoring




Cost of acquisition Moderate



Availability and status Still in R&D stage, not used in practice yet


Recommended for further testing Yes, follow up of TNO TISNET

Recommended for use Yes, can also be done in limited pilot

Other comments Take into account huge dataflow


D03031/CG9/0M0/001999/ep ARCADIS 101

5.8 Radiometric Fingerprinting

General

Feedback from users Medusa Explorations is currently using gamma-

spectrometry to detect polluted pavements

Advantages Quick scan of pavement compositions based

on known isotopic fingerprints


Disadvantages Integrating over depths of 20-30 cm, therefore

not always able to distinguish top layer

compositions

Technical scoring







Availability and status Available in The Netherlands, still partially R&D


Recommended for further testing No, not at this moment (see also comment below)

Recommended for use Only when more is known about pavement

composition and wearing processes in the future.

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 102

5.9 Hyperspectral

General

Feedback from users None received

Advantages Quick scan of pavement composition and

condition


Disadvantages Cores are still needed to check the exact

nature of the problem

Combined cracking and ageing of pavement

may give incorrect results, more study

needed

Technical scoring







Availability and status Still in R&D stage, TRB is currently testing hyper

spectral monitoring techniques for Non-

Destructive Testing (NDT) of pavements


Recommended for further testing No, not at this moment (see also comment below)

Recommended for use Only when more is known about pavement

composition and wearing processes in the future.

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 103

5.10 Asphalt Paving Research & Innovation (ASPARI)

General

Feedback from users First results seem promising

Advantages Provides the art of paving with a solid

scientific base

Done during paving process, no hindrance

Disadvantages Quality of technique can only be proven

when the first pavement defects occur, this

may be 5-10 years from now

Complex and indirect

Technical scoring




Cost of acquisition Moderate, estimated 20,000€ per vehicle



Availability and status Still in R&D stage, not used in practice yet



Recommended for use No, although contractors are interested in the

methods and techniques

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 104

5.11 Tagging (RFID)

General

Feedback from users Received feedback from ARCADIS NL Asset

Management for successful application in

buildings and airplane industry

Advantages Quick access to up to date information

Possibility to retrieve and store inspection

results at the spot

Disadvantages May not be possible to place or reach RFID

tags in all structure elements

Still need to setup complete chain, from

design and build to maintenance (DBM)

Technical scoring




Cost of acquisition Expensive



Availability and status Commercially available. Used in other industries.



Recommended for use Yes, include DBM chain

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 105

5.12 Magnetostrictive sensing

General

Feedback from users No feedback received from users

Advantages Capable of detecting and reporting the

location of corrosion damage

Fast, A 100 m long suspender rope can be

inspected in as little as 15 minutes

Disadvantages No continuous monitoring

Still a need to visit the structure

Technical scoring



Speed of acquisition Slow




Availability and status Commercially available. Scientists used MsS to

inspect the suspenders on the George Washington

Bridge (GWB), Bridge of the Americas in Panama,

the Verrazano- Narrows Bridge, the Bronx

Whitestone Bridge, and the Throgs Neck Bridge in

New York City.


Recommended for further testing No, since already commercially available

Recommended for use Yes, maybe for the limited number cable bridges

in the Netherlands

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 106

5.13 Digital photography

General

Feedback from users None received

Advantages Archiving structure (deterioration) history

Bridges with no visible defects on photo may

be skipped in expert inspection planning

Disadvantages Can still only see the outside of a structure

element

Inspections by experts still needed

Technical scoring







Availability and status Technique is available. R&D in UK.




Other comments


D03031/CG9/0M0/001999/ep ARCADIS 107

5.14 In-situ sensors

General

Feedback from users Live demonstration of ASR monitoring in A59

structures was given at Koenders Instruments HQ

Advantages Continuous monitoring

Quality data on loads, temperatures,

moisture, joint expansion, etc.

Internet based monitoring enables quick alert

when a sensor fails

Disadvantages Multiple sensors are needed to monitor one

structure, one for moisture, one for cracks,

one for stress, etcetera.

Technical scoring




Cost of acquisition Very expensive



Availability and status Commercially available. Used in practice, e.g. A59

ASR monitoring



Recommended for use Yes, for cost-benefit reasons, best to be used in

structures with a long life-span

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 108

5.15 SmartPaint

General


Advantages Can indicate the loss of protective function of

anti-corrosive paint layers

Disadvantages So far C-Cube SmartPaint only works on steel

Technical scoring







Availability and status Commercially available. Used in practice


Recommended for further testing No, already under investigation

Recommended for use Yes

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 109

5.16 Fibre optics

General


Advantages Very sensitive technique to measure

deformation, temperature and strain

Using FBG’s problem locations can be

identified

Early warning

Disadvantages May be costly and complicated to integrate in

existing constructions

Technical scoring




Cost of acquisition Expensive



Availability and status Commercially available in Netherlands. Still R&D


Recommended for further testing No, already tested in IJkdijk


Other comments


D03031/CG9/0M0/001999/ep ARCADIS 110

5.17 Deformation sensors

General


Advantages Very sensitive technique to measure

deformations

Warning for overloading of structures

Possible to integrate in existing constructions

Disadvantages Installation and setup cost may be high

Technical scoring




Cost of acquisition Moderate, depending on life expectancy of the

structure



Availability and status Commercially available in Netherlands and UK




Other comments


D03031/CG9/0M0/001999/ep ARCADIS 111

5.18 InSAR

General

Feedback from users No feedback received from users

Advantages Rapid surveying

Accurate, deformation rates are measured as

mm/year

Disadvantages Two images must be accurately co-registered

to a sub-pixel level to ensure that the same

ground targets are contributing to that pixel

Another source of error present in

interferograms is caused by the propagation

of the waves through the atmosphere

Technical scoring







Availability and status Commercially available in Netherlands, R&D


Recommended for further testing No, some studies are already done on an ad-hoc

basis

Recommended for use No, still R&D many questions still unanswered

before deciding about implementation

Other comments

http://en.wikipedia.org/wiki/Image_registration


D03031/CG9/0M0/001999/ep ARCADIS 112

5.19 Mobile 3D Laser Mapping

General

Feedback from users Joost Swarts (ARCADIS NL) showed positive

results using this system

Advantages Rapid surveying at driving speed

Accurate, 3D point cloud

Objects can be measured on screen

Disadvantages Measuring objects in point clouds is still done

manual, however, software is improving fast

Not suitable in rain or mist

Obstruction by other vehicles

Technical scoring



Speed of acquisition Extremely rapid, post-processing: 2 days/day

Cost of acquisition Cheap: 100€ per kilometre (including processing)



Availability and status Commercially available in the Netherlands


Recommended for further testing Yes, in particular to test automated feature

extraction

Recommended for use Yes

Other comments


D03031/CG9/0M0/001999/ep ARCADIS 113

ANNEX 6 Review & evaluation meeting

Advanced Inspections Project

Review & Recommendations Meeting

16th September 2009, 2pm – 4pm

C139, van de Burghweg 1, Delft

Attendees:

Aad de Winter (DVS, RWS) (AW)

Rosalind Reid (DVS, RWS & HA) (RR)

Martijn Koster (DVS, RWS) (MK)

Albert Manenschijn (DVS, RWS) (AM)

Fons Bots (DI, RWS) (FB)

Ton Maagdenberg (DVS, RWS) (TM)

Wim Koops (PIM, RWS) (WK)

Henk Kersten (PIM, RWS) (HK)

Jos Bakker (Arcadis) (JB)

Sander Borghuis (Arcadis) (SB)

Apologies:

Jaap Bakker (DI, RWS)

Erik Vos (DI, RWS)

Guido Debeus (DVS, RWS)

1. Project Overview

RR: explained the purpose of the meeting and gave an overview of the Advanced

Inspections Project (reference presentation slides).

SB: presented the inspection techniques investigated during the project. The scoring of

techniques was explained and those techniques which appeared most promising (reference

presentation slides).

JB: explained issues relating to the further development and embedment of new techniques

within RWS (reference presentation slides).

2. Discussion on findings and recommendations

AM: it is difficult to harvest the benefits from techniques that produce better information

about the condition of assets because of the other steps in the maintenance process (for

example, budget allocation process taking a long time).

TM: a research project is already underway on the causes of pavement ravelling. DVS are

already investigating the use of lasers line scanners to detect ravelling.

TM: agrees with the report findings that little is known about changes in pavement

composition in relation to the wearing processes. Pavement composition measurement

technologies are only of interest when more is known about the composition behaviour in

relation to wearing.


D03031/CG9/0M0/001999/ep ARCADIS 114

MK: techniques for improved structural inspections need to be able to be added to existing

structures.

AW: asked what techniques seemed promising for future development.

FB: for structures, sensors are useful to develop further. Sensors that can tell more about

what is happening at the heart of a construction are most valuable, for example, small

deformations in steel and concrete. Sensors would need to be applied to existing structures

using risk-based analysis. It is less important to fit sensors to new constructions as they do

not have the same problems as older structures.

AW: asked what the first step in development should be.

FB: the first step for future development is to do a best practice pilot (similar to IJkdijk?),

maybe on the A27 for hydrogen brittleness. There is already some monitoring with in-situ

sensors in place, for example for ASR. This could be expanded further.

AM: at this stage, not keen to recommend development of a particular technology. It is

useful to now have an inventory of possible new techniques through the Advanced

Inspections Project. Further investigation is needed on the maintenance philosophy and

how best to work with the market (e.g. through DBFM contracts). Would like to see some

scenarios e.g. if all contracts were DBFM in the future. Need to consider how we stimulate

innovations. Consider what we expect in the future (in terms of new technology) and how

we functionally specify our requirements to the market. RWS are buyers and therefore

should steer and control and also need to remain knowledgeable about possible new

techniques. Would expect the next step to be taken forward through the IPW program.

AW: asked if car-as-sensor technique, which can collect data at a higher frequency, would be

useful for early detection of ravelling.

MK: Ravelling is a typical Dutch problem. Information on ravelling isn’t needed more

frequently than is currently collected (i.e. once/twice every year). This is partly because it

takes a long time to plan maintenance work (several months) so having more frequent data

doesn’t help much. RWS need enough data to predict their four year program of work. The

laser line scanner is currently under investigation by RWS/TNO so should not be identified

as an ‘innovation’ in the report.

AM: for winter pavement problems that happened this year, sensors that detect these

problems early would be useful. Road users were also involved in alerting RWS to

problems through a telephone number. Taking preventative measures could have reduced

the impact of the problems. RWS need to know more about the costs and benefits of taking

preventative maintenance and consider different scenarios.

AW: enquires whether a relatively cheap ‘early warning’ system for ravelling would be

welcome or not.

TM: Once ravelling starts it is difficult to take action to stop it getting worse.


D03031/CG9/0M0/001999/ep ARCADIS 115

RR: Business cases that compare costs and benefits of enhanced inspections and

preventative maintenance with current inspections and corrective maintenance, would help

to evaluate if it is worth investing in improved inspection techniques that enable

preventative maintenance. This cost comparison would need to be over a long period, say

5/10 years. This would help to test whether having better asset knowledge reduces

maintenance costs in the longer term.

AM: agreed this would be useful. This would help decide whether it is worth investing in

techniques like sensors.

RR: asked about the role of OTC in developing business cases like this.

WK: explained that once OTC is established it would be involved in business improvement

initiatives. OTC could be involved in developing and implementing a new technique but

first needed to know what was being implemented and who in RWS was commissioning it.

WK also asked who in RWS would be deciding what should be implemented. RWS should

be acting as a client.

MK: There are existing projects underway to make improvements in the next 2 – 3 years.

Innovation projects need to focus on the longer-term future. We should be seeking to learn

from other industries such as medical, clothing, aeronautics. Ideas need to be tested in labs

and investigated with Universities. In the report, availability in 2-3 years should not be a

criteria when scoring techniques.

JB: explained that in the ranking of the most promising new techniques, higher weighting

was given to those techniques that could be implemented in the next 2-3 years. If this is not

as important then weighting factor would be amended in the report.

AW: explained that innovations often stop at implementation because users are not

involved at the early stage.

WK: explained that OTC helped address this through functionally testing processes with

stakeholders. When IPW is ready to test and implement something then this should be

done with OTC.

AW: the approach is to decide on the most promising new techniques, understand more

about the costs/benefits and then involve stakeholders and OTC.

RR: Strong business cases are increasingly important in the Highways Agency. In order to

progress new developments, business cases should be produced showing the costs/benefits

of the technique and costs/benefits for different approaches to maintenance (e.g.

preventative).

WK: Once more detailed business cases are in place, OTC could have a role in development

and implementation.

advanced inspection project final report_tcm174-283988

Documents

spaceborne techniques

hyperspectral techniques

current practice

leading practice

pavement condition

project background

project focus

monitoring of defects