Final_Report_2.pdf

Identification of Effective Next Generation Pavement Performance Measures and Asset Management Methodologies to Support MAP-21 Performance Management Requirements Draft Final Report PUBLICATION NO. XX September 2016

U.S. Department of Transportation Federal Highway Administration ii

FOREWORD

Transportation Asset Management (TAM) is a process that enables agencies to use engineering and business practices to make more informed and transparent investment decisions. Several processes are required for successful TAM practices, including effective decision analysis methodologies that are informed by quality data and performance measures. The interaction between models, data and performance measures is a critical one for the success of TAM implementation. This report documents the development of a new generation of pavement performance measures, as well as a set of methodologies that are designed to support decisionmaking within a TAM framework.

This report seeks to answer the following questions: How should transportation department measure and assess the condition of their pavement network? What processes are required for successful TAM implementation? How should tradeoff analysis be implemented in departments of transportation? Each of these questions are addressed throughout this report, and the recommendations in this report represent a step forward in terms of both TAM and performance management.

Cheryl A. Richter, Ph.D., P.E., Director Office of Infrastructure Research and Development

Notice

This document is disseminated under the sponsorship of the U.S. Department of Transportation in the interest of information exchange. The U.S. Government assumes no liability for the use of the information contained in this document. This report does not constitute a standard, specification, or regulation.

The U.S. Government does not endorse products or manufacturers. Trademarks or manufacturers’ names appear in this report only because they are considered essential to the objective of the document.

Quality Assurance Statement The Federal Highway Administration (FHWA) provides high-quality information to serve Government, industry, and the public in a manner that promotes public understanding. Standards and policies are used to ensure and maximize the quality, objectivity, utility, and integrity of its information. FHWA periodically reviews quality issues and adjusts its programs and processes to ensure continuous quality improvement.

iii

TECHNICAL DOCUMENTATION PAGE

1. Report No.

XX

2. Government Accession No.

3. Recipient’s Catalog No.

4. Title and Subtitle Identification of Effective Next Generation Pavement Performance Measures and Asset Management Methodologies to Support MAP-21 Performance Management Requirements: Draft Final Report

5. Report Date September 2016

6. Performing Organization Code

NONE

7. Author(s) James Bryce, Gonzalo Rada, Sam Van Hecke, Joseph Zissman

8. Performing Organization Report No.

9. Performing Organization Name and Address Amec Foster Wheeler Environment & Infrastructure, Inc.

12000 Indian Creek Court, Suite F Beltsville, MD 20705

Cambridge Systematics, Inc.

100 CambridgePark Drive, Suite 400 Cambridge, MA 02140

10. Work Unit No. (TRAIS)

11. Contract or Grant No.

DTFH61-15-C-00044

12. Sponsoring Agency Name and Address U.S. Department of Transportation Federal Highway Administration Office of Asset Management 1200 New Jersey Avenue, SE Washington, D.C. 20590

13. Type of Report and Period Covered Draft Final Report September 2015 – September 2016

14. Sponsoring Agency Code

15. Supplementary Notes The Contracting Officer’s Representative (COR) was Nadarajah Sivaneswaran, HRDI-20

16. Abstract This report documents research designed to develop a next generation set of performance measures for pavements, as well as develop a process that supports the comparison of many disparate assets in support of full implementation of Transportation Asset Management (TAM) within Departments of Transportation (DOTs). TAM is a process that enables DOTs to use engineering and business practices to make better decisions in a more transparent manner, the methods presented in this report contribute to enhancing decisionmaking within a DOT. First, the current state of practice for pavement performance measures is documented, and, based on the findings from a literature review and discussions with many DOTs, a set of performance measures is developed and recommended for adoption by DOTs. Then, the current state of practice for TAM implementation is reviewed, and it was found that methods for implementing comprehensive tradeoff analysis were in need of further development and enhancement. Based on the review of the current state of TAM, an approach to conducting comprehensive tradeoff analysis is presented in this report. Finally, this report includes detailed work plans designed to validate the recommended measures and tradeoff analysis methodologies.

17. Key Words Performance Measures, Pavements, Performance Management, Asset Management, Tradeoff Analysis

18. Distribution Statement No restrictions.

19. Security Classif. (of this report) Unclassified

20. Security Classif. (of this page) Unclassified

21. No. of Pages

22. Price iv

Form DOT F 1700.7 (8-72)Reproduction of completed page authorized

SI* (MODERN METRIC) CON

APPROXIMATE CONVERSIO

Symbol When You Know Multiply By

LENGTH

in inches 25.4 ft feet 0.305 yd yards 0.914 mi miles 1.61

AREA

in2 square inches 645.2 ft2 square feet 0.093 yd2 square yard 0.836 ac acres 0.405 mi2 square miles 2.59

VOLUME

fl oz fluid ounces 29.57 gal gallons 3.785 ft3 cubic feet 0.028 yd3 cubic yards 0.765

NOTE: volumes greater than 1000 L

MASS

oz ounces 28.35 lb pounds 0.454 T short tons (2000 lb) 0.907

TEMPERATURE (exac oF Fahrenheit 5 (F-32)/9 or (F-32)/1.8

ILLUMINATIO

fc foot-candles 10.76 fl foot-Lamberts 3.426

FORCE and PRESSURE lbf poundforce 4.45 lbf/in2 poundforce per square inch 6.89

APPROXIMATE CONVERSIO

Symbol When You Know Multiply By

LENGTH

mm millimeters 0.039 m meters 3.28 m meters 1.09 km kilometers 0.621

AREA

mm2 square millimeters 0.0016 m2 square meters 10.764 m2 square meters 1.195 ha hectares 2.47 km2 square kilometers 0.386

VOLUME

mL milliliters 0.034 L liters 0.264 m3 cubic meters 35.314 m3 cubic meters 1.307

MASS

g grams 0.035 kg kilograms 2.202 Mg (or "t") megagrams (or "metric ton") 1.103

TEMPERATURE (exac oC Celsius 1.8C+32

ILLUMINATIO

lx lux 0.0929 cd/m2 candela/m2 0.2919

FORCE and PRESSURE N newtons 0.225 kPa kilopascals 0.145

*SI is the symbol for th International System of Units. Appropriate rounding should e (Revised March 2003) v

TABLE OF CONTENTS

CHAPTER 1. INTRODUCTION

BACKGROUND

OBJECTIVES

PROJECT APPROACH

REPORT ORGANIZATION

CHAPTER 2. EXISTING PAVEMENT PERFORMANCE MEASURES

BACKGROUND

DEFINING PERFORMANCE MEASURES

PAVEMENT PERFORMANCE MEASURES IN A SYSTEMS BASED

CONTEXT

CHAPTER 3. NEXT GENERATION PAVEMENT PERFORMANCE MEASURES

BACKGROUND

EVALUATION OF THE RECOMMENDED MEASURES

DISCUSSION

CHAPTER 4. NEXT GENERATION ASSET MANAGEMENT TRADEOFF

METHODOLOGIES

BACKGROUND

ESTABLISHING LINKS BETWEEN INVESTMENT AND AGENCY

OBJECTIVES

IDENTIFYING THE RIGHT MIX OF FUNDING BETWEEN BRIDGES,

PAVEMENTS, AND OTHER ASSETS

SETTING ASSET CONDITION AND PERFORMANCE TARGETS

DEVELOPING LEAST LIFE-CYCLE COST STRATEGIES

MITIGATING RISK THROUGH INVESTMENT

DISCUSSION

CHAPTER 5. RECOMMENDED METHODOLOGIES TO FACILITATE

COMPREHENSIVE TRANSPORTATION ASSET MANAGEMENT

BACKGROUND

TRADEOFF ANALYSIS OVERVIEW

RECOMMENDED APPROACH

METHODOLOGIES TO ENHANCE THE RECOMMENDED APPROACH

METHODOLOGIES TO IMPROVE PROJECT HARMONIZATION

DISCUSSION

CHAPTER 6. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

SUMMARY

CONCLUSIONS

RECOMMENDATIONS

REFERENCES

APPENDIX A. WORK PLAN TO FURTHER DEVELOP AND VALIDATE THE

CONCEPTUAL STRATEGIC PAVEMENT PERFORMANCE MEASURES

BACKGROUND

vi

PERFORMANCE MEASURES

VALIDATION ACTIVITIES

DELIVERABLES, EXPECTED SCHEDULE AND LEVEL OF EFFORT

REFERENCES

APPENDIX B. WORK PLAN TO FURTHER DEVELOP AND VALIDATE THE

RECOMMENDED ASSET MANAGEMENT TRADEOFF ANALYSIS

METHODOLOGIES

BACKGROUND

TRADEOFF ANALYSIS OVERVIEW

THE TWO-STEP APPROACH TO TRADEOFF ANALYSIS

VALIDATION OBJECTIVES AND TASKS

EXPECTED SCHEDULE AND LEVEL OF EFFORT

REFERENCES

vii

LIST OF FIGURES

Figure 1. Diagram. Systematic process demonstrating the role and development of performance measures Figure 2. Diagram. Mapping performance measures to a systems framework Figure 3. Graph. Assumed roughness deterioration as a function of age Figure 4. Graph. Assumed transverse cracking growth as a function of age Figure 5. Graph. Assumed rut growth as a function of age Figure 6. Graph. Probability of each pavement being eligible for preservation Figure 7. Graph. Roughness prediction for sampled pavement segments Figure 8. Graph. Transverse cracking prediction for pavement segments Figure 9. Graph. Fatigue cracking prediction for sampled pavement segments Figure 10. Graph. Rutting prediction for sampled pavement segments Figure 11. Chart. Comparison of (a) calculating good-fair-poor measures using the NPRM criteria, and (b) estimating treatment categories using RSI approach Figure 12. Graphic. Traditional approach to tradeoff analysis Figure 13. Graphic. Bottom-up approach to tradeoff analysis Figure 14. Graphic. Top-down approach to tradeoff analysis Figure 15. Graphic. Recommended two-step approach to tradeoff analysis Figure 16. Graphic. Corridor harmonization in tradeoff analysis Figure 17. Graphic. Applying a geographic bonus to projects in similar areas Figure 18. Graphic. Integrated delay costs viii

LIST OF TABLES

Table 1. Responses to the criteria weighting survey based on respondent grouping Table 2. Effect of different perspectives on the RSI Table 3. Good, fair and poor criteria for pavements Table 4. Good-fair-poor criteria from NPRM. (4) Table 5. General characteristics of the HPMS sample used in the example Table 6. Example pavement ‘A’ from HPMS estimated performance using HERS models Table 7. Example pavement ‘B’ from HPMS estimated performance using HERS models Table 8. Example pavement ‘C’ from HPMS estimated performance using HERS models ix

LIST OF ACRONYMS AND SYMBOLS

AASHTO - American Association of State Highway Transportation Officials

AC - Asphalt concrete

BCR - Benefit Cost Ratio

CCI - Critical Condition Index

DOTs - Departments of Transportation

ESALs - Equivalent Single-Axis Loads

EUAC - Equivalent Uniform Annual Cost

FAST - Fixing America's Surface Transportation

FHWA - Federal Highway Administration

FWD - Falling Weight Deflectometer

HERS - Highway Economics Requirements System

HPMS - Highway Performance Monitoring System

IHS - Interstate Highway System

IMPP - Interstate Maintenance Preservation Program

IRI - International Roughness Index

LCCA - Life cycle cost analysis

LLCC - Lowest life cycle cost

LOS - Level-of-service

LTPP - Long-Term Pavement Performance

LRTP - Long Range Transportation Plan

M&R - Maintenance and Rehabilitation

MAP-21 - Moving Ahead for Progress in the 21st Century Act

MEPDG - Mechanistic-Empirical Pavement Design Guide

NAPCOM - National Pavement Cost Model x

NCHRP - National Cooperative Highway Research Program

NHS - National Highway System

NPV - Net Present Value

PCI - Pavement Condition Index

PCC - Portland Cement Concrete

PCR - Pavement Condition Rating

PHT - Pavement Health Track

PMS - Pavement Management System

R&R - Reconstruction and Rehabilitation

RQFS - Road Quality Forecasting System

RSI - Remaining Service Interval

RSL - Remaining Service Life

SHA - State Highway Administration

SOC - Simple Option Cost

STIP - Statewide Transportation Improvement Plan

TAM - Transportation Asset Management

TAMP - Transportation Asset Management Plan

VMT - Vehicle Miles Traveled

ES-1

EXECUTIVE SUMMARY

ES-2

ES-3

ES-4

CHAPTER 1. INTRODUCTION

BACKGROUND

Transportation Asset Management (TAM) is a process that enables departments of transportation (DOTs) to use engineering and business practices to make more informed and transparent investment decisions. (1) An important aspect of successful TAM implementation within a DOT is a clear relationship between the performance of assets classes and allocation of resources among asset classes. Maturing this relationship requires that asset performance be clearly defined and tied to quantitative measures in such a way that enables DOTs to embrace data driven decisionmaking, to track progress towards goals and to develop performance targets. However, developing the relationship between performance and resource allocation requires that the performance of an asset be clearly defined, and measures be developed in support of monitoring this performance. Additionally, once the relationship between asset performance and resource allocation is defined for each asset, DOTs must develop a means to compare the consequences of investing in multiple, disparate types of assets.

This report documents research to develop a next generation set of performance measures for pavements, as well as to develop a process that supports the full implementation of a comprehensive approach to TAM. The development of performance measures and tradeoff analysis methodologies were coupled in recognition that the two concepts are integrally connected; effective tradeoff analysis requires robust performance measures. In addition, interest in improving performance measures and asset management methodologies has been solidified by guidance provided in the Moving Ahead for Progress in the 21st Century (MAP-21) Act and the Fixing America's Surface Transportation (FAST) Act. (2, 3)

Many pavement performance measures can be found throughout literature, and these measures are currently implemented across DOTs. (4) However, most of the existing measures are limited to describing the current physical condition of one component of a pavement system from a single perspective. Essentially, the majority of existing pavement performance measures do not describe the pavement from a systems based perspective, and instead only describe one component of the pavement system. Furthermore, the majority of current performance measures are tactical (i.e., not strategic) and reactive. Therefore, the condition measured by the current generation of pavement performance measures is due to past decisions, and does not inform how decisions made today affect the future performance of pavements. Accordingly, this research recommends performance measures that take a strategic and systematic perspective on pavement performance.

Performance measures directly inform tradeoff methodologies, and thus the development of performance measures must address how they can be used in cross-asset resource allocation and tradeoff analysis. Cross-asset resource allocation offers an opportunity for decisionmakers to perform tradeoff analysis while evaluating investment options during a period when needs exceed available resources. The National Cooperative Highway Research Program (NCHRP) Report 806 provides one taxonomy of allocation approaches. (5) While much work and insight has gone into the development of the current generation of methodologies, they maintain some shortcomings. One of the most notable is the lack of a feedback loop that describes the lost benefits of delaying particular projects. The methodology developed and presented in this report implements this feedback loop using response curves.

OBJECTIVES

The goal of this project was to improve the overall state of performance management by developing more strategic pavement performance measures, as well as developing and recommending tradeoff methodologies to support asset management decisionmaking. The specific objectives of the research documented in this report were to:

1. Identify or conceptually develop more strategic performance measures, and their means of collection, in order to strengthen performance management and better link investments to long-term performance. Consideration was to be given to how the data supporting the next generation method would be collected, analyzed, and presented and the potential costs and benefits of using this type of performance measure.

2. Identify or conceptually develop methodologies to enable full implementation of a comprehensive asset management plan, including trade-off analysis from a common ground among disparate assets that are traditionally individually assessed and managed.

PROJECT APPROACH

The two distinct objectives to this project were researched in parallel. The recommended approaches for comprehensive TAM helped define the recommended performance measures and the performance measures in turn helped define the TAM tradeoff analysis recommendations.

The overall approach to developing and recommending a set of pavement performance measures was to first conduct a thorough literature review. The literature review addressed the following:

• What is the definition of a pavement performance measure, the main purposes of the performance measures, and the criteria against which the measures can be evaluated?

• What performance measures are currently used for pavements in both domestic and international transportation agencies, and what are the main attributes of these measures?

• How have performance measures been developed for other complex systems throughout domestic and international literature?

Through conducting the literature review, a definition of pavement performance measures was adapted for this project, and a set of criteria were developed that can be used to evaluate the overall effectiveness of performance measures. In addition, it was found that other fields, particularly environmental management, define performance measures in terms of systems based frameworks. A simple systems based framework developed by the Organization for Economic Cooperation and Development (OECD) was presented as a tool for use in developing the next generation of pavement performance measures.

The approach to developing and recommending asset management tradeoff methodologies also began with a literature review, albeit with a different focus than the literature review for pavement performance measures. The literature review for tradeoff methodologies focused on:

• Identifying the current state of practice among DOTs

• Assessing the strengths and drawbacks of many current cross-asset resource allocation techniques

• Understanding how each of the components of tradeoff analysis are currently implemented in practice

The literature review resulted in identifying three main approaches to conducting tradeoff analysis; traditional approaches (allocation based on historical levels of funding for each asset independently), top-down approaches, and bottom-up approaches.

Following the literature review, phone interviews were held with five State DOTs using a targeted set of questions. Each DOT provided input about their unique experiences with both pavement performance measures and approaches to implementing comprehensive TAM practices. The information from the literature review was supplemented with the information gathered from the interviews with the State DOTs in an effort to understand why certain concepts found throughout literature either were or were not incorporated in the field.

The results from the literature reviews and phone interviews, along with additional work that will be described throughout this report, were used to develop a set of recommended pavement performance measures and a recommended tradeoff analysis approach. The performance measures were developed based on representing the pavement as a system and then defining performance measures as the outcome of the system. The tradeoff methodologies were developed by identifying the strengths in current asset management practice and leveraging them within a robust cross-asset resource allocation technique.

REPORT ORGANIZATION

This report is organized into the following six chapters and two appendices:

• Chapter 1 provides background information, details the project objectives, summarizes the project approach and presents the report organization.

• Chapter 2 presents background information that supports the development of a set of pavement performance measures, including the development of a system-based framework to define pavement performance.

• Chapter 3 presents the recommended performance measures, as well as outlines and demonstrates the calculation and interpretation of the measures.

• Chapter 4 reviews literature and current practices for TAM implementation in domestic and international agencies, as well as provides the background information to support the recommendations made in Chapter 5.

• Chapter 5 recommends and details a methodology to support full implementation of asset management, including trade-off analysis from a common ground among disparate assets that are traditionally individually assessed and managed.

• Chapter 6 presents a project summary as well as the major conclusions and recommendations that resulted from the project.

• Appendix A contains a self-contained work plan that can be used to validate the pavement performance measures recommended in this report.

• Appendix B contains a self-contained work plan that can be used to validate the asset management tradeoff methodologies recommended in this report.

CHAPTER 2. EXISTING PAVEMENT PERFORMANCE MEASURES

BACKGROUND

NCHRP Report 551 defines performance measures as indicators that can be used to measure progress towards a goal. (6) NCHRP Report 708 defines performance measures as quantifiable criteria used to measure progress towards goals. (7) In this same light, Amekudzi and Meyer defines performance measures as indicators of system effectiveness and efficiency. (8) Simpson et

al. discussed pavement performance measures as ways to consistently assess the functional condition of pavements. (4) Finally, section 6028 of the newest congressional funding bill, known as the FAST Act, carries forward the performance management requirements from MAP-21, and it discusses performance measures in the context of performance management. In other words, performance measures are used to communicate current condition and measure progress towards specified goals.

A comprehensive and succinct definition of performance measures, that includes each aspect of the previously discussed definitions, was presented in NCHRP Synthesis 311. (9) The definition was taken from the US Federal Highway Administration (FHWA) National Performance Review, and it reads:

“Performance measurement is a process of assessing progress toward achieving predetermined goals, including information on the efficiency with which resources are transformed into goods and services (outputs), the quality of those outputs (how well they are delivered to clients and the extent to which clients are satisfied) and outcomes (the results of a program activity compared to its intended purpose), and the effectiveness of government operations in terms of their specific contributions to program objectives.” (pg. 5)

Performance measures play a critical role in the proper management of pavements, and can significantly influence long-term strategies regarding resource allocation. The role of performance measures has been thoroughly discussed in many sources, such as in the NCHRP Project 20-74, which highlights the importance of performance measures for communicating physical conditions of assets. (1) The role of pavement performance measures in the resource allocation process is both to develop maintenance and rehabilitation (M&R) strategies, and to benchmark the benefits of the M&R strategies. In other words, performance measures are not only used to determine which roads to maintain in a given year, but they must be responsive to the application of M&R in the current year, as well as provide a means to communicate the effect of particular M&R strategies over time. Therefore, effective performance measures should give an indication of the current M&R needs for a pavement, and how the performance of the pavement over time is affected by different M&R strategies.

Within performance management and performance based planning, performance measures are the link between the decisions and the outcomes. With this view, Grant et al. described performance measures as serving the following five purposes: (10)

1. To clarify the definition of goals – A performance measure is a tool that is used in converting broad goals into measurable objectives.

2. To monitor or track performance over time – Metrics are used to track performance on regular basis (e.g., yearly, monthly).

3. As a reference for target setting – Metrics are used as the basis for selecting a target that is intended to be achieved.

4. As a basis for supporting policy and investment decisions by comparing alternative options – Metrics are used as a basis for comparing alternative investments or policies in order to make decisions.

5. To assess the effectiveness of projects and strategies – Metrics are what enable measurement to assess whether projects and strategies have worked to further goals.

The five purposes described by Grant et al. for performance based planning are similar to the six objectives described by Haas et al. for performance objectives. (11) These six objectives are:

1. Assessment of physical condition (in terms of level of service provided to road users, such as smoothness, structural distress such as cracking, structural adequacy such as the capacity to carry a specified number of loads, surface friction as related to safety, and various other [usually qualitative] condition measures related to culverts, signs, fences, curb and gutter, etc. such as good, fair or poor).

2. Determination of asset value, that can vary with accounting base (e.g., financial or management accounting) and with valuation method.

3. A monitoring mechanism for assessing policies in terms of their effectiveness and/or compliance with predefined policy objectives.

4. Provision of information to users or customers.

5. Use as a resource allocation tool in terms of quantifying the relative efficiency of investments across competing alternatives.

6. Diagnostic use for early identification of accelerated deterioration of assets and for appropriate remedial actions.

Several different approaches can be taken to develop pavement performance measures. However, the approach to developing the performance measures in each case should reflect the expected use or applications of the measures. For example, McGhee describes the development of performance measures at the Virginia DOT, and noted that the goal of the measures is to communicate the current condition of the pavement as succinctly and accurately as possible. (12) On the other hand, Christiansen et al. describes the process of first adopting a set of performance models for Nordic countries, and then selecting a set of performance measures based on these models, along with a number of other factors. (13) In the case of McGhee, it is considered important to communicate the current condition of the pavement whereas, while in the case of Christiansen et al., it is considered important to develop measures based on those that can be used to predict future conditions.

DEFINING PERFORMANCE MEASURES

Based on a review of relevant literature, a definition of performance measures was adopted for this project as follows: pavement performance measures are indicators that are used to communicate the current condition of pavements and facilitate the measurement of progress towards specified goals. (4, 6, 7, 8, 9) Implicit in this definition is the evaluation of the current condition of the asset, as well as long term trends in asset condition and an assessment of decisions made to achieve specific goals for the asset. In this sense, performance measures should have two fundamental characteristics:

• Ability to be used as a communication tool to inform stakeholders of the condition of the pavement network at a given point in time relative to specified goals.

• An analytical tool to connect resource allocation decisions and maintenance policies to outcomes (i.e., long-term pavement condition as a function of decision-making).

Current State of Pavement Performance Measures

A thorough review of literature was conducted in an effort to identify pavement performance measures that have been developed to date. It was found that pavement performance measures could be classified into several categories based on the information that the measure was designed to communicate. For example, several measures were developed to communicate the surface condition of the pavement, whereas others were developed to communicate the structural condition of the pavement. This section of the report summarizes the findings of the literature review on pavement performance measures.

Composite Pavement Condition Measures

Composite measures combine individual condition measures (e.g., cracking, rutting, etc.) into proxy values that are designed to represent the overall condition of the pavement. Composite measures are expected to provide the benefit of succinclty summarizing the condition of the pavement into a single number that is easily interpretable. Although the methods used to calculate composite performance measures differ considerably from each other, the basic approach to rating a pavement using a composite measure can be summarized by the following four steps:

1. Assign each pavement the best possible score, representative of a pavement with no distresses (e.g., a Pavement Condition Index (PCI) of 100).

2. Measure the type, extent and severity of the distresses present on the pavement.

3. Translate the measured distresses into a scaled value (e.g., a deduct value) and combine the deduct values.

4. Combine the orginial score from step 1 with the scaled values in step 3 to arrive at a composite measure of pavement condition.

Papagiannakis et al. synthesized the various performance measures used throughout the States, and found that the uniqeness of how each State defines and calculates pavement condition is a limiting factor when making comparisons between each State. (14) An early and widely used composite performance measure is the PCI developed by the US Army Corps of Engineers for use in the PAVER pavement management system (PMS). The PCI, which is described by ASTM

International, is an index calculated using each type and severity of distress present on the pavement. (15) A score of 100 (i.e., perfect condition) is first assigned to the pavement, and then specific vlaues are deducted from the score based on the distress type and severity present on the pavement. Many modern performance measures are based on the PCI methodology.

More examples of similar composite performance measures include:

• Mississippi DOT version of the Pavement Condition Rating (16)

• Florida DOT Pavement Condition Rating (17)

• Illinois DOT Condition Rating Survey (18)

• Texas DOT Condition Score (14)

• Iowa DOT Pavement Condition Index (19)

• Virginia DOT Critical Condition Index (12)

The Indiana DOT measured the performance of the pavement network in terms of the Pavement Condition Rating (PCR) up through the year 2010. The PCR was based on distress data collected from annual condition surveys. However, an analysis of the quality of data collection revealed that errors in the distress measurements resulted in errors in the PCR calculations of ± 20 percent, and that the PCR could not discern differences early in the performance life of the pavement. (20) These errors in the PCR have the potential to lead to significant differences in how M&R needs are identified for a given pavement segment.

It is important to note that, although many of the composite condition measures share the same name, significant differences may exist regarding how each measure is calculated. For example, the Ohio DOT’s PCR does not include roughness whereas the Florida DOT’s PCR does include roughness. In addition, State DOTs will typically calibrate deduct values to match the needs of the specific department.

Measures Based on Individual Pavement Surface Distress Metrics

In addition to composite measures, individual pavement distresses have been used as performance measures in several contexts. For example, several State DOTs define pavement condition using the International Roughness Index (IRI). (21) Given that individual pavement condition measures are not scaled to similar ranges, the extent and severity of these distresses are generally translated into a relative scale. An example of translating distresses into a relative scale is by defining ranges for each measurement within which a pavement can be classified as either in good, fair or poor condition.

Simpson et al. thoroughly outlined the use of cracking, rutting (for asphalt pavements), roughness and faulting (for jointed concrete pavements), and validated these measures using condition data collected specifically for the project in question. (4) In addition, Simpson et al.

made several recommendations regarding the use of the individual distresses as performance measures in the FHWA Highway Performance Monitoring System (HPMS). (4) An example of the recommendations from Simpson et al. are:

• Roughness (ride quality) should be measured using the IRI, summarized over 0.1-mile (160 m) segments and interpretted as:

o Good: IRI less than 95 inches/mile (1.50 m/km) o Fair: IRI between 95 inches/mile (1.50 m/km) and 170 inches/mile (2.68 m/km) o Poor: IRI greater than 170 inches/mile (2.68 m/km)

The FHWA issued a Notice of Proposed Rulemaking (NPRM) in January 2015 to establish performance measures to assess the condition of the pavements on the National Highway System (NHS) and Interstate Highway System (IHS). (22) The condition of the pavements is to be determined based on the IRI, rutting, cracking percent, and faulting. This resulted in FHWA proposing four pavement performance measures to assess pavement condition as follows:

• Percentage of pavements on the IHS in Good condition

• Percentage of pavements on the IHS in Poor condition

• Percentage of pavements on the NHS (excluding IHS) in Good condition

• Percentage of pavements on the NHS (excluding IHS) in Poor condition

Measures Based on Pavement Structural Condition Metrics

As described in Park et al., pavement condition is generally described in terms of either functional or structural condition. (23) The functional condition describes the ability of the pavement to provide a specified level of service, and it is typically measured using the pavement roughness. On the other hand, the structural condition describes the ability of the pavement to carry the anticipated traffic load throughout its design life. In other words, the structural condition influences the rate of change of the functional condition throughout the pavement life.

Pavement structural capacity is generally measured non-destructively using deflection measurement devices.

Some examples of structural condition measures for pavements are:

• The structural adequacy index (SAI) was an early index developed by Haas et al. for use with the Benkelman Beam, but can be used for any measure of deflection with appropriate calibration. (24) The SAI is defined on a scale from 1 to 10, and a maximum tolerable deflection (MTD) can be established based on the pavement and number of expected equivalent singel axle loads (ESAL’s). Deflections that match the maximum tolerable deflection would be considered a 5, the worst deflection would be considered a 0, and the minimum deflection would be considered a 10.

• The Texas DOT Structural Capacity Index (SCI) is a network-level index that is based on the American Association of State Highway Transportation Officials (AASHTO) effective structural number of the pavement. (25) The SCI is determined based on falling weight deflectometer (FWD) testing results along with ESAL data and total pavement thickness. The SCI can be described as the ratio of the effective structural number of the pavement to the required structural number, where the required structural number is documented in a table as a function of the resilient modulus of the subgrade and ESALs.

The SCI is only applicable to flexible pavements.

• Bryce et al. developed a structural index for use in the Virginia DOT PMS known as the Modified Structural Index (MSI), which is based on FWD testing results, ESAL data and total thickness of the pavement. (26) Simialr to the SCI, the MSI was developed for use on flexible pavements, and it is essentially the ratio of the effective structural number of the pavement to the required structural number.

• Bryce et al. developed a structural measure for composite (asphalt over Portland cement concrete (PCC)) pavements in the State of Virginia that is based on FWD testing results.

(27) The measure was calibrated by relating the performance ratio of pavements, defined as the actual measured performance divided by the expected performance, to the deflection basin charateristics of the pavement. In essence, if the performance ratio is less than one, the pavement is said to be structurally inadequate.

Much of the work on developing structural capacity measures has been completed using static deflection measuring devices (e.g., the FWD). However, recent research has shown traffic speed deflection measurment devices to be promising technologies for measuring the structural capacity of a pavement network. (28, 29) These devices include the Rolling Wheel Deflectometer (RWD), and the Traffic Speed Deflectometer (TSD). The TSD is the focus of a current demonstration project, Transportation Pooled Fund Study TPF-5(282)1, where the device is being demonstrated at the network-level for many State DOTs.

Cost Based Measures

Several measures of cost have been proposed as performance measures, especially at the network and strategic levels in pavement management. For example, Proctor et al. proposed the asset sustainability index (ASI). (30) The ASI is defined as the ratio of the amount budgeted for highway infrastructure preservation divided by the amount needed to adequately sustain infrastructure at a targeted condition over the long term. Some additional examples of cost based performance measures include:

• The Ohio DOT pavement sustainability ratio and sustainability gap. (30) The former is similar to the ASI, while the latter is defined as the budget shortfall between the pavement needs and actual budget.

• The debt index was proposed in the NCHRP Report 551 as a cost based performance measure. (6) The debt index is defined as the ratio of deterioration or lost value (for a single pavement or a pavement network) to its replacement value.

1 http://www.pooledfund.org/Details/Study/518

• The Metropolitan Transportation Commission, the planning organization for the San Francisco Bay Area region, which includes 109 cities and counties in Northern California, uses deferred M&R costs as a performance measure. (31) A multi-year M&R program is developed, and deferred M&R is calculated as the cost of M&R not selected in a given year due to budget constraints. Essentially, deferred M&R costs are the same as the sustainability gap used by the Ohio DOT.

• The Washington DOT calculates two measures related to cost: the historical discounted equivalent uniform annual cost (EUAC), and the ESAL efficiency factor. (32) In addition to these, the Washington DOT also evaluates the backlog of M&R needs by calculating the amount of M&R that was deferred in past years due to budget shortcomings. The key performance measure used within the Washington DOT pavement management system is the cost per lane-mile year, which is the EUAC normalized by the segment length. This performance measure is particularly well suited for analyzing areas of the pavement network that have historically high costs in order to identify potential causes and mitigate future high costs. In essence, the measure used by the Washington DOT is a proxy of the performance of the pavement and the M&R decisions made. Well performing pavements that are well managed will tend to have lower EUAC per lane-mile values than poor performing or poorly maintained pavements.

• Costello et al. proposed the use of a vehicle operating cost index as a performance measure for the New Zealand Ministry of Transport. (33) This measure is effectively a combination of the pavement roughness, vehicle mix and the vehicle-kilometers traveled into a representative combined cost for vehicles travelling along a pavement. The increased costs due to pavement roughness were estimated using the Highway Development and Maintenance Management System (HDM) 4 models.

Remaining Service Interval

A performance measure used by many DOTs to describe the condition of their pavement network is the remaining service life (RSL) of the pavement. In an effort to update the remaining service life concept, Elkins et al. proposed the remaining service interval (RSI). (34) Fundamentally, the remaining service life concept applied to pavements limits the information required to make good decisions because it does not adequately capture the fact that more optimal pavement management consists of treatment strategies over multiple points in time. The RSI concept was proposed as a way to describe pavement treatment intervals (i.e., multiple numbers) as opposed to describing a pavement threshold interval (i.e., one number).

The FHWA recently completed a project aimed at validating the RSI concept at project and network levels, and demonstrated its use as a potential performance measure. (35) As part of the RSI validation process, the FHWA team is demonstrating optimal treatment strategies for several pavement segments based on lowest life cycle costs (LLCC). Although the research is still ongoing, the use of the RSI as a performance measure is currently being demonstrated.

Essentially, the RSI as a metric that describes treatment strategies is a next step in using timing as a performance measure.

Other Measures

Performance measures are designed to reflect the strategic goals of a DOT. This opens the possibility of several types of performance measures to be developed based on a DOTs strategic direction. For example, several DOTs have adopted friction measures as a proxy for pavement related safety considerations. Some DOTs have prioritized sustainability, and thus have developed or adopted performance measures related to environmental sustainability. Some examples of these other measures include:

• Burati et al. developed a guide for DOT use to develop acceptance plans and quality acceptance plans. (36) As part of this work, Burati et al. reviewed several acceptance characteristics, or measures that can be used to determine whether the construction and material quality conforms to set standards. One of the measures discussed in Burati et al.

is the percent within limits, which is a value that estimates the percentage of the tested section that falls within a set band of acceptable values. (36) The percent within limits can be calculated for multiple characteristics (e.g., void content and compaction), and related to expected performance, and can be used as a construction quality based performance measure.

• Nishio et al. describes performance indicators in Japan, and includes rate of achievement of nighttime noise limits under the objective of improving the environment. (37)

Pavements play a significant role in roadway noise, and certain pavement characteristics (e.g., surface type) are strongly correlated to roadway noise. (38)

• As a safety related pavement performance measure, the Maryland State Highway Administration (SHA) routinely measures pavement friction statewide and reports the results in terms of the friction number. (39) Pavement friction has been shown to be directly related to the expected number of crashes along the pavement segment. (40) Thus, the measured friction can be compared to levels of acceptable pavement friction as a benchmark of pavement safety considerations.

• NCHRP Project 08-74, as documented in NCHRP Report 708, developed a guidebook for sustainability related performance measures for use by DOTs. (7) One important finding by Zietsman et al. is that many measures currently tracked by State DOTs can be useful to also measure progress towards sustainability related goals and objectives. In addition, Zietsman et al. notes that sustainability based performance measures require moving from output based measures to outcome based measures. (7) In other words, although performance measures should be mapped to sustainable goals, many measures can be developed based on currently collected data.

• Several point based systems have been developed for use as sustainability performance measures, where the total number of points represents the level of sustainability. For example, Greenroads is a point based rating system that is designed to quantify sustainable practices within the design and construction of roadways. (41) Another sustainability tool that can be viewed as a point based performance measure is INVEST (Infrastructure Voluntary Evaluation Sustainability Tool) that was developed by the

FHWA. (42)

Criteria for Assessing Performance Measures

Pavement performance measures are designed to capture and to communicate the physical condition of pavements, as well as to provide useful metrics for decision-making. As such, the selection of performance measures should involve a thorough evaluation of how each measure compares to specified criteria. As is documented in NCHRP Report 666, the criteria for selecting measures often include the following: (43)

• Feasibility

• Policy sensitivity

• Ease of understanding

• Usefulness in decision-making

Grant et al. laid out the following factors as being important for selecting measures: (10)

• Does it represent a key concern? The measure that is selected will have important implications on strategies that are selected. Moreover, the selection of measures should address key issues, accounting for factors such as risk, importance to the public, and implications for policy.

• Is it clear? Is the measure understandable to policy makers, transportation professionals, and the public?

• Are data available? Transportation staff need to consider the feasibility and practicality to collect, store, analyze data and report performance information for the selected measures.

• Can it be forecasted? Are there realistic methods to compare future alternative projects, investment approaches, or strategies using the measure?

• Is the measure something the DOT and its investments can influence? A good measure does not need to be something that a DOT directly controls (e.g., a DOT cannot directly dictate vehicle miles travelled (VMT)). At the same time, it will be important to select measures that can be influenced through policy and investment decisions in order for the measure to be useful in supporting investment decision-making (e.g., a DOT can influence VMT through its decisionmaking).

• Is the measure meaningful for the types of services or area? While consistency in metrics can be valuable, it is also important to make sure that a measure is meaningful to the area or system to which it is applied.

• Improvement direction is clear. In some cases, DOTs choose measures but do not state clearly whether they desire the measure to increase or decrease, which is particularly problematic when the measure could be interpreted differently depending on one’s perspective.

According to the US Army Corps of Engineers, characteristics of a good performance measure include: (44)

• Meaningful: significant and directly related to the mission and goal

• Responsibility linked: matched to an organizational unit responsible for achieving the measure

• Organizationally acceptable: valued by those within the organization

• Customer focused: reflect the point of view of the customers and stakeholders

• Comprehensive: include all key aspects of performance

• Balanced: include several types of measures, i.e., outcome, efficiency, and quality measures

• Timely: use and report data in a reasonable time-frame

• Credible: based on accurate and reliable data

• Cost effective: based upon acceptable data collection and processing costs

• Compatible: integrated with existing financial and operational systems

• Comparable: useful for making comparisons with other data over time

• Simple: easy to calculate and interpret

Finally, as stated in the NPRM, the FHWA used the following criteria to assess the proposed measures for national use and readiness for implementation accurately and reliably in order to carry out the National Highway Performance Program: (22)

• Is the measure focused on comprehensive performance outcomes?

• Has the measure been developed in partnership with key stakeholders?

• Is the measure maintainable to accommodate changes?

• Can the measure be used to support investment decisions, policy making and target establishment?

• Can the measures be used to analyze performance trends?

• Has the feasibility and practicality to collect, store, and report data in support of the measures been considered?

• Does the measure exhibit good timeliness, consistency, completeness, accuracy and accessibility, and can it be integrated within DOTs data systems?

Many more resources exist that document the important criteria for a performance measure.

However, each of the criteria defined in the previously cited work are representative of the evaluation criteria found throughout the literature that was reviewed.

In addition to assessing criteria from literature, phone interviews were conducted with State DOTs in an effort to understand the criteria that each deem important for performance measures.

Pavement and asset management personnel from five State DOTs were interviewed in order to discuss the characteristics they view as important for performance measures. They are: Vermont DOT, Arizona DOT, Indiana DOT, Maine DOT and Kansas DOT. Some of the main findings from the interviews include:

• Each of the State DOTs identified the importance of performance measures as a communication tool. For example, the personnel from the Kansas DOT identified the need to specifically perform work within their State to identify a performance measure with which they…