42 JANUARY 2017 | JOURNAL AWWA • 109 :1 | BOULOS

Globally, aging water and wastewater infrastructure is rapidly deteriorating and sometimes failing, with potentially dire human, environmental, and financial consequences. As this aging process continues, utilities are faced with costs they can no longer afford. Recent advances in smart water network (SWN) modeling tech-

nology have played a crucial and growing role in addressing this challenge. SWN technology has equipped utilities with a comprehensive set of decision-making tools designed to help them effectively manage these critical assets while main-taining the desired level of service at the lowest life-cycle cost. By seeking the lowest life-cycle cost and reinvesting the savings in their infrastructures, utilities are able to help close the infrastructure gap and place their assets on the road to efficiency and sustainability. These advances have the capacity to transform how infrastructure is conceived, designed, and delivered to enable a sustainable future and continue to promote economic development and quality of life.

SMART WATER NETWORK

MODELING TECHNOLOGY

CAN HELP UTILITIES

EFFECTIVELY MANAGE

CRITICAL INFRASTRUCTURE

ASSETS WHILE

MAINTAINING THE DESIRED

LEVEL OF SERVICE AT THE

LOWEST LIFE-CYCLE COST.

infrastructure

PAUL F. BOULOS

Optimal Scheduling of Pipe Replacement

2017 © American Water Works Association

BOULOS | 109 :1 • JOURNAL AWWA | JANUARY 2017 43

ASSET MANAGEMENT OVERVIEWWater and wastewater infrastruc-

ture in the United States is continu-ously aging, and current capital spending is not expected to keep pace with needs. More than one mil-lion miles of water pipes are nearing the end of their useful lives, meaning they need to be replaced sooner rather than later (AWWA 2012). Assuming every water pipe is replaced, the cost could reach more than $1 trillion over the next couple of decades (Boulos & Wiley 2013). Total capital investment needs for US wastewater systems are estimated at $298 billion over the same period.

With between 700,000 and 800,000 mi of public sewer mains, pipes represent the largest capital need, making up three-quarters of total needs (Boulos & Walker 2015). Yet this critical infrastructure is pro-jected to see the largest investment gap, falling 73% short of needs. Because of insufficient investment, this gap will only widen with time as more pipes reach the end of their effective service lives. Compounding this problem is the need to address stricter regulatory requirements, growing populations, increased ser-vice demands, limited water supplies, increased potential for flooding and drought, and decreased state and federal funding. This does little to assuage the public’s health and eco-nomic needs or maintain its confi-dence in the safety of our drinking water and the ecological health of the nation’s waterways.

A key objective in managing aging assets is to balance system perfor-mance and cost (USEPA 2013, Shamir & Howard 1979). Selection of main segments for replacement should be based on a comparison of two alternatives: (1) replacing the pipes and incurring the replacement (capital) costs and future marginal (operations plus maintenance plus risk) costs associated with the new lines or (2) keeping them for the time being, saving the replacement costs but incurring future marginal costs. The conventional end-of-design life

(e.g., 50-, 75-, or 100-year) main replacement cycle, while easy to communicate to the public and poli-cymakers, can lead to replacing mains that have years of remaining useful life and does not take into account the risk of not replacing the mains (Flynn 2014). It is therefore essential to base any replacement strategy on a risk-based economic life cycle rather than short-term ben-efits such as the pipe’s initial capital cost, end-of-service life when the asset can no longer serve its intended purpose, or end-of-physical life when the asset becomes physically non-functioning. Managing water and wastewater pipes to the lowest

life-cycle cost and accounting for fail-ure risk (i.e., the product of failure likelihood and failure consequence) is the best basis for creating an optimal replacement strategy and reducing the infrastructure gap (Figure 1).

The cost of owning any asset (in this case, pipe) warrants its replace-ment when it exceeds the new asset’s (pipe’s) ownership cost (Figure 2). This serves as the basis for strategic asset management. Strategic asset management is aimed at maintaining the desired level of service at the lowest life-cycle cost—the best appropriate cost for rehabilitating or replacing an asset. The goal is to find the point in the asset’s life cycle or

Co

st

Time

Pipe replacement needs

SWN

Availablefunding

FIGURE 1 Closing the water and wastewater infrastructure gap

SWN—smart water network

Co

st

Pipe Age

FIGURE 2 Planning pipe replacement

Optimal

New pipeownership cost

Existing pipeownership cost

New pipeownership cost

Existing pipeownership cost

<

>

2017 © American Water Works Association

economic life where the cost of replacement is balanced against the expected cost (or consequences) of failure, the accelerating cost to main-tain it, and the declining level of ser-vice. To find this point, a utility must know which pipes are most critical to sustain desired system performance and service level and then gauge their remaining useful lives. Critical pipes are those that have a high failure risk and major consequences if they fail. Consequence of failure can depend on a number of pipe attributes, including buried depth, number of connected customers, and proximity to critical facilities (e.g., rivers, wet-lands, hospitals, schools, manufactur-ing, high-density housing, airports, military facilities). The remaining use-ful life of these pipes is estimated using probabilistic pipe-deterioration-condition curves. This helps utilities make risk-based decisions by choos-ing the right mains for replacement at the right time. This process will help reduce pipeline failures and their adverse effects, minimize life-cycle

costs, and give stakeholders the assur-ance of long-term system sustainabil-ity and security.

If the goal is to minimize the pres-ent value of the total cost of owner-ship, the optimum period for replac-ing mains represents the end of the economic life for an existing pipe, determined as the point in time when the present value of the future total ownership cost is no longer the low-est life-cycle cost alternative (i.e., when it is cheaper to buy a new pipe). This enables utilities to ensure the smartest distribution of dollars spent on replacement and renewal of their assets. It also helps them create reliable, cost-effective, risk-based capital improvement plans that enhance operational efficiency and quality of service.

SWN MODELS Recent advances in SWN modeling

technology can assist utilities in addressing the challenges related to aging water and wastewater infra-structure. SWN models can be

divided into four general categories: geographic information system (GIS), network, asset integrity, and finance models. Used together, these highly complementary models constitute a decision support and strategic asset management tool for utility managers (Figure 3). By using them to conduct a wide-ranging risk-based economic life-cycle cost analysis, they can esti-mate the lifetime performance of each individual pipe in their water and wastewater networks, including their expected failures, repairs, and even-tual replacement, as well as associ-ated costs of optimizing the manage-ment of those high-risk assets. This knowledge can help utilities preserve structural, hydraulic, and water qual-ity integrity and ensure that their pipe networks operate well into the future at maximum cost savings.

GIS. A GIS model acts as the central storehouse for all water and waste-water network asset data (e.g., pipe length, diameter, material type, loca-tion, age, surrounding soil condition, leakage/break/defect history). From

FIGURE 3 Example GIS-centric, risk-based life-cycle cost pipe replacement model

GIS—geographic information system

44 JANUARY 2017 | JOURNAL AWWA • 109 :1 | BOULOS

2017 © American Water Works Association

this spatial database, inputs to all models are generated and model results are associated for query, anal-ysis, and display on the network map. Over time—as pipes are rehabilitated, repaired, or replaced—asset mainte-nance, break history, and inspection data are updated. This allows utilities to develop a condition assessment and rating system for all pipes. The data can also be used to predict the future rate of deterioration and esti-mate when a pipe will fail. The ability to view this asset analysis data on a map is a powerful tool for recogniz-ing spatial trends and hot spots.

Network. Network models repre-sent the most effective way to pre-dict network behavior of water and waste water systems under a range of loading and operating conditions. They make use of continuity, energy and momentum principles, and reaction kinetics to track actual sys-tem hydraulic and water quality performance over time in meeting desired levels of service. The expected levels of service help char-acterize the importance of each asset and its criticality. These assets include fire-fighting and storage capacity, disinfectant residual con-centration, system losses, inflow and infiltration, extent of flooding and overflows, frictional losses, and operating pressures. They are shaped by minimum performance requirements such as regulations, permit conditions, water quality standards, and discharge limits. Through their criticality analysis capabilities, these models can also carry out vulnerability assessments to determine the consequence of failure of each pipe.

Asset integrity. Asset integrity mod-els estimate the failure potential for all network pipes. As a pipe ages and deteriorates, the likelihood of failure increases, thereby increasing risk. The objective is not necessarily to deter-mine the cause of pipe failures, but rather when they might be expected to fail in the future. Because it is not fea-sible to directly observe buried (under-ground) infrastructure, condition

assessments for water and wastewater mains are normally based on their esti-mated remaining useful lives. Using existing pipe condition inspection and historical failure rate data, condition curves of future failure or survival probability can be determined. The

most widely used models to fore-cast pipe failure include the non-homogeneous Markov chain, non-homogeneous Poisson process, linear extended Yule process, Cox propor-tional hazards model, and time-based probabilistic Weibull and Herz models. The accuracy of the models is determined by how well pipe cohorts are grouped in terms of common fac-tors. Critical pipes (i.e., those with the highest risk of failing) are identified, and those with the greatest negative impact are given the most attention.

Finance. Finance models balance capital (replacement) expenditure against marginal cost to minimize the overall cost of asset ownership (Figure 4). Marginal cost can be expressed as the present value of the expected risk cost (or consequences)

of pipe failure, the accelerating cost to maintain it (maintenance cost), and declining level of service (opera-tional cost). These costs can also include internal resources and over-heads; leakage, additional pumping, and traffic and disruption costs; and costs related to sustainability, resil-iency, greenhouse gas emissions, and health and safety. At the same time, the discounted (or present value) replacement cost declines as pipe renewal is deferred. The total owner-ship cost, or expected life-cycle cost,

By seeking the lowest life-cycle cost and

reinvesting the savings in their infrastructures,

utilities are able to help close the infrastructure

gap and place their assets on the road to

efficiency and sustainability.P

rese

nt V

alu

e

Age to Replacement

FIGURE 4 Optimal timing of pipe replacement

Replacement(capital) cost

Total Ownership Cost

Optimal Marginal cost(operations +

maintenance + risk)

Arrow indicates where the present value compared with the age to replacement is not optimal, i.e., where the marginal cost and replacement cost are equal, not when the combined impact is least.

BOULOS | 109 :1 • JOURNAL AWWA | JANUARY 2017 45

2017 © American Water Works Association

typically forms a convex shape, whereas the minimum point (i.e., when the gradient is zero) depicts the optimal time of replacement. This optimal replacement time represents the pipe’s economic life. Knowing the full costs and revenues generated by

water and wastewater systems enables utilities to determine a long-term funding strategy and allocate utility resources in the most efficient way.

CONCLUSIONSWater and wastewater utilities

worldwide face the constant chal-lenge of sustaining and improving the performance and extending the life of their assets with limited resources. SWN models can play a growing role in meeting this chal-lenge because of their inherent abil-ity to optimize scheduling of pipe replacement while maintaining desired levels of service. These mod-els allow utilities to proactively reha-bilitate or replace their pipes on a continual basis rather than wait to repair failing or damaged ones when the move is considerably more expensive and disruptive to system operations. They can also serve as a logical framework for making func-tional industry procurement changes from a low-bid first costs procure-ment strategy (i.e., lowest design and construction costs) to a life-cycle costing that will result in the most cost-effective, efficient, and resilient infrastructure. This framework is aligned with the American Public Works Association’s call for estab-lishing life-cycle costing of public infrastructure as a replacement for low-bid procurement (Ross 2001).

A critical part of any capital improvement plan is gaining the

support of administrators, elected offi-cials, ratepayers, and other nontechni-cal stakeholders. Implementing such recommendations is expensive, and decision-makers demand transpar-ency and supporting documentation. With a sound SWN framework,

utilities are better positioned to receive the support they need to main-tain their financial and operational sustainability well into the future.

EDITOR’S NOTEAn abridged version of this article

appeared in the December 2016 issue of Opflow.

REFERENCESAWWA, 2012. Buried No Longer: Confronting

America’s Water Infrastructure Challenge. AWWA, Denver.

Boulos, P.F. & Walker, A., 2015. Fixing the Future of Wastewater Systems With Smart Network Modeling. Journal AWWA, 107:4:72. http://dx.doi.org/10.5942/jawwa.2015.107.0057.

Boulos, P.F. & Wiley, A., 2013. Can We Make Water Systems Smarter? Opflow, 39:3:20. http://dx.doi.org/10.5991/OPF.2013.39.0015.

Flynn, R., 2014. Replacing the Right Main at the Right Time. Water Online, Dec. 5, 2014. www.wateronline.com/doc/replacing-the-right-main-at-the-right-time-0001 (accessed Aug. 22, 2016).

Ross, D.A., 2001. Infrastructure Research Needs: A Public Works Perspective. First Annual Conference on Infrastructure Priorities: A National Infrastructure Research Agenda, Oct. 24–26, Washington.

Shamir, U. & Howard, C.D.D., 1979. An Analytic Approach to Scheduling Pipe Replacement. Journal AWWA, 71:5:248.

US Environmental Protection Agency (USEPA), 2013. Primer of Condition Curves for Water Mains. National Risk Management Research Laboratory,

Office of Research and Development, EPA/600/R-13/080, Washington.

ABOUT THE AUTHORPaul F. Boulos is the president, chief operating officer, and chief innovation officer of Innovyze Inc., 370 Interlocken

Blvd., Broomfield, CO 80021 USA; paul.boulos@ innovyze.com. He has coauthored more than 200 articles and 10 books on water and wastewater engineering, available in the AWWA Bookstore (www.awwa.org/store).

http://dx.doi.org/10.5942/jawwa.2017.109.0002

Knowing the full costs and revenues generated by

water and wastewater systems enables utilities to

determine a long-term funding strategy and

allocate utility resources in the most efficient way.

Journal AWWA welcomes comments and feedback at journal@awwa.org.

AWWA RESOURCES• Asset Management Resource

Community. www.awwa.org/resources-tools/water-knowledge/asset-management.aspx.

• Asset Management Program Development Case Studies on Collection, Storage and Utilization of Asset Information. Cameron, A., 2016. AWWA Annual Conference Proceedings. AWWA Catalog No. ACE_0084169.

These resources have been supplied by Journal AWWA staff. For information on these and other AWWA resources, visit www.awwa.org.

46 JANUARY 2017 | JOURNAL AWWA • 109 :1 | BOULOS

2017 © American Water Works Association