Slide 1

A Tutorial on Establishing Effective Work Zone Performance MeasuresTracy Scriba, Federal Highway AdministrationandGerald Ullman, Texas Transportation Institute

What is Work Zone Performance Measurement?Determining progress or level of attainment towards specific defined agency objectives regarding:SafetyMobilityCustomer satisfactionAgency/contractor efficiency and qualityEvidence based

Why are Work Zone Performance Measures Important and Useful?Quantifies work zone impactsEffects on motoristsActions taken to mitigate impactsEffects of actions taken

Why are Work Zone Performance Measures Important and Useful?Guides investment decisions, policy development, and program prioritiesIdentify trendsRefine policies and proceduresEmphasize accountability

Why are Work Zone Performance Measures Important and Useful?Assists in communication and outreachAids agencies in telling their storyFacilitates better public understanding and acceptanceEnsures that elected officials have correct information

6WZ Rule: WZ Data and PMsRule Goal: Better understand and manage WZ impactsRelated Provisions:Impacts AssessmentUse of DataProcess ReviewsMove practice forward

6The WZ Rule is built on better understanding and managing WZ safety and mobility impacts. This requires data and information.

Recognizing this, the WZ Safety and Mobility Rule includes provisions for impacts assessment, data, and process reviews. In updating this Rule, FHWA expanded the data provision to include operational data. Operational data such as traffic volumes/exposure and queues are critical to understanding both the safety and mobility impacts of WZs.

Having WZ performance measures can helps an agency focus in on an agreed upon set of key parameters so that data collection efforts can be done efficiently and data used effectively to make decisions. 7WZ Rule Data ProvisionsRequires agencies to use WZ data at both the project and process-levels to manage and improve work zone safety and mobilityProject level: Improvements while projects are underwayProcess level: Broad improvements over timeRecommends that agencies maintain data and information resources to support the use of WZ data for the above activities 7Work Zone Performance Nationally: Work Zone Self AssessmentDone annually since 200310.3 average (0 to 15 scale)Challenges in Measuring Work Zone PerformanceDefining objectives that are:AcceptableRealisticUsefulIdentifying and obtaining useful dataAllocating sufficient resources (staff and money) Work Zone Performance Measurement MythsMeasurements must be obtained for all work zonesReality: It may be only necessary to monitor a set of key work zones Measurements must be continuous at the work zones that are monitoredReality: It may only be necessary to obtain a few key measurements during the right times and at the right locations in a work zoneWork Zone Performance Measurement MythsA lot of time and money must be spent on data collectionReality: Depending on project characteristics, low-cost data collection methods may be available and appropriate

Work Zone Performance Measurement MythsA lot of time and money must be spent on data collectionReality: depending on project characteristics and agency goals, low-cost data collection methods may be available and appropriate

Work zone performance measurement is simply a tool that can help an agency accomplish its mission more effectivelyMaking Work Zone Performance Measurement Happen in your Agency

Key StepsIdentify and engage stakeholdersSelect appropriate measuresIdentify available data sourcesDefine analysis requirementsAssign roles and responsibilitiesDefine methods for disseminating resultsPeriodically review and refine measures1. Identify and Engage Stakeholders to Define:Needs for performance measuresPerformance goals or targets (based on policy and procedures)Performance measurement champions

2. Select the MeasuresBase on the performance goals or targets identifiedKeep to a manageable numberConsider data needsStart simple, refine and expand as needs and uses dictateMeasures to considerExposureVehicle exposureWork activity exposureSafetyCrashes or accidents Safety surrogatesMobilityQueues Travel times or delaysTravel time reliabilityAgency ratingsCustomer complaints or ratings

% days or night when work activity occursAverage hours of work per day or night% work hours with lane closuresLane-mile-hour of closures (overall, during certain time periods)

Counts (possibly by severity)% of crashes of a certain typeRates per vehicle-mile-traveled (or change in rate)% of projects that exceed an acceptable crash rate (or change in rate)$ or % change in crash costs compared to the expected no-work-zone costsHighway worker injuries or rate per hours worked

173. Identify Data Needs and SourcesExposure dataProject plansAgency construction management databaseInspector diariesLane closure request databasesAutomatic traffic recorder (ATR) countsPlanning division AADT estimatesManual countsReal-time electronic traffic surveillance dataSafety dataStatewide crash databaseAgency collected work zone crash dataOccupational safety recordsAgency field reviewsService patrol or EMS dispatch logs

3. Identify Data Sources (contd)Mobility dataElectronic traffic surveillance technologiesObservations and documentation of queues by field personnel or transportation management center operations staffTravel time runs by staff or internsAgency inspection scoresCustomer survey responsesCustomer complaint files

Electronic Traffic Surveillance SourcesPoint measurements of speed, volumeExisting transportation management system spot sensorsWork zone ITS deploymentsPortable traffic monitoring devicesPoint-to-point travel timesAutomatic vehicle location, identification systemsLicense-plate recognition systemsCellular telephone signal trackingBluetoothPurchase of private-sector dataManual Documentation of Queues

4. Define Analysis RequirementsEstimating travel times and delays from queue length documentationEstimating queue lengths from spot sensor data

Travel Times from Queue LengthsNotes:Appropriate only if pre-work zone conditions do not experience queuesOther speed/density/flow relationships could be assumed

Queues from Spot Speed SensorsSteps:Divide roadway into segments of uniform speedExamine speeds and volumes hour-by-hourCompare speeds across sensorsSum regions where speeds are below thresholdsCompute individual and vehicle-hours of delay

5. Assign Roles and ResponsibilitiesData collectionAnalysisDecision-makingDissemination

6. Define Methods for Disseminating ResultsConsider key usersTechnical managers and decision makersPublic and political leadersTailor messages to each user group as appropriate

7. Review and Refine MeasuresKey questions:Do measures aid in evaluating progress?Do measures point to changes that may be needed in policies or procedures?

Work Zone Performance Measurement Getting Started

One ExampleAn agency currently reviews its total fatal work zone crashes each yearThe Agency Selects a Measure:Change in severe (injury + fatal) crash rates from pre-work zone conditionsInitial focus will be on freeway work zonesEach of 4 districts to evaluate two of its work zones annuallyResultsDistrictSevere Crash Rate BeforeSevere Crash Rate AfterChange12.7/100 mvm4.6/ 100 mvm3.5/100 mvm4.3/100 mvm+30%-7%21.2/100 mvm1.5/100 mvm1.3/100 mvm2.5/100 mvm+8%+67%31.7/100 mvm0.9/100 mvm2.4/100 mvm1.9/100 mvm+41%+111%43.1/100 mvm1.6/100 mvm3.0/100 mvm2.1/100 mvm-3%+31%Observations made by the agencyRates do trend higherIncreases are highly variableAgency performs detailed analysis of projectsProject durations and volumes vary widelyIdentifies some good practices, issues to addressThey also modify their analysis methodologyDistrictSevere Crashes ExpectedaSevere Crashes That OccurredDifference13.619.1526+1.4+6.921.110.4012-1.1+1.6335.44.9483+12.6-1.9426.63.2404+13.4+0.8Overall104.3138+33.7(+32%)a Expected crashes computed using Highway Safety Manual approachCrash Analysis Results the Following YearAnother ExampleAn agency has adopted a policy that no work zones will create more than a 15-minute delay for motorists9 complaints per month from motorists Are delays occurring, and are they greater than 15 minutes?Institute procedures for inspectors to document all queues that occur and estimate delays

Results (out of 31 projects)InspectorQueuesPer Vehicle DelayTotal Delay15 daysLength: Ave = 0.4 miles, Max = 3.1 miDuration: Ave = 2.1 hrs, Max = 4 hrsAve = 7.6 minMax = 58.9 min 2394 veh-hrs21 dayLength: ave = 1.4 miles, max = 1.8 miDuration: 1.5 hrsAve = 2.0 minMax = 2.8 min75 veh-hrs315 daysLength: Ave = 0.8 miles, max = 1.2 miDuration: Ave = 1.5 hrs, Max = 2.0 hrsAve = 1.4 minMax = 2.1 min840 veh-hrs

ObservationsImpacts being observed at a small (< 10 %) number of projectsUnclear how many motorists are being excessively delayedAdjust monitoring to determine % of vehicles experiencing greater than 15 minutes of delay

Recomputed Results InspectorQueues (Speeds < 35 mph)Per Vehicle Delay% Vehicles with > 15 min DelayTotal Delay15 daysLength: Ave = 1.4 mi Max = 3.1 mi

Duration: Ave = 2.1 hrs Max = 4 hrsAve = 7.6 minMax = 58.9 min 1.2 % of total ADT

15.1 % of queued vehicles2394 veh-hrsNext steps taken:Interns are hired to assist in traffic monitoring (requested by project engineers)Requirement to document queues is extended to the maintenance sectionPercent of projects that exceed agency delay threshold is added to dashboard metrics on agency website

A Final ExampleThe director of an agency assigns the traffic operations division to establish a work zone performance measurement program1. Identify and Engage StakeholdersKey agency staffDivision staff District engineersPublic information officeProject design divisionPerformance targetsVehicle delays < 15 min 85% of users who encounter a delay warned in advanceCrashes no more than 20% greater than expected without a work zone present2. Select Appropriate MeasuresIndividual vehicle delays during hours of delayPercent of customers indicating awareness of when and where delays were going to occurRatio of actual severe crashes to expected crashes 3. Identify Data SourcesProject designers: identify any projects with peak-hour volume-to-capacity > than 0.9Interns: conduct peak-hour travel time studies at projects after each phase changeInspectors: document queues during any short-term lane closures in Districts3. Identify Data Sources (contd)University: conduct annual survey of customer opinions about work zonesStatewide crash database to be used to evaluate 5 high-volume projects annually4. Define Analysis RequirementsInterns to gather and analyze project inspector data on queues monthlyInterns compute delays based on queue dataSafety division to provide expected crash frequencies at selected project locations (based on Highway Safety Manual procedures)University customer survey analysis5. Assign Roles and ResponsibilitiesDivision staffHiring and supervising internsCoordinating information collection and analysis on key projects Assessing implications of measures on agency policies and operationsDistrict engineersIssue requirement regarding project inspector documentation of queues

5. Roles and Responsibilities (contd)Project design division staffIdentify key projectsSafety division staffDevelop baseline crash frequencies Public information officeCreation of summary document for directorPosting results on agency website (dashboard)6. Define Methods for Disseminating ResultsExecutive summary documentDirector, public officialsDashboard summary on websiteGeneral publicPresentation of results at annual short course7. Periodically Review and Refine MeasuresRevisit every two years as part of process assessmentStakeholders to meet to discuss need to expand or contract effort, identify additional information needs or issues to be examinedResourcesA Primer on Work Zone Safety and Mobility Performance MeasurementWork Zone Performance Measures Pilot TestDomestic Scan on Work Zone Assessment, Data Collection, and Performance Measurement

Available at http://www.ops.fhwa.dot.gov/wz/decision_support/performance-development.htm

Questions?Tracy Scriba, FHWA(202) 366-0855Tracy.Scriba@dot.gov

Gerald Ullman, TTI(979) 845-9908g-ullman@tamu.edu