National Academies Press: OpenBook

Guidance for Calculating the Return on Investment in Transit State of Good Repair (2019)

Chapter: Appendix B - Annotated Bibliography

« Previous: Appendix A - Literature Review Summary
Page 93
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 93
Page 94
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 94
Page 95
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 95
Page 96
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 96
Page 97
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 97
Page 98
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 98
Page 99
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 99
Page 100
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 100
Page 101
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 101
Page 102
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 102
Page 103
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 103
Page 104
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 104
Page 105
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 105
Page 106
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 106
Page 107
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 107
Page 108
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 108
Page 109
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 109
Page 110
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 110
Page 111
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 111
Page 112
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 112
Page 113
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 113
Page 114
Suggested Citation:"Appendix B - Annotated Bibliography." National Academies of Sciences, Engineering, and Medicine. 2019. Guidance for Calculating the Return on Investment in Transit State of Good Repair. Washington, DC: The National Academies Press. doi: 10.17226/25629.
×
Page 114

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

B-1 A P P E N D I X B Annotated Bibliography Models of State of Good Repair Investment Spy Pond Partners, LLC; KKO & Associates, LLC; Harry Cohen; and Joseph Barr. TCRP Report 157: State of Good Repair: Prioritizing the Rehabilitation and Replacement of Existing Capital Assets and Evaluating the Implications for Transit. Transportation Research Board of the National Academies, Washington, D.C., 2012. TCRP Report 157 presents a framework for transit agencies to use for prioritizing capital asset rehabilitation and replacement decisions. By applying this framework, a decision maker can answer questions about asset rehabilitation and replacement investment decisions. The published report is accompanied by four Microsoft Excel models, which are available electronically via the TRB website. This report and the models will be a valuable resource for transit agencies and will be of interest to regional, state, and federal agencies that oversee, plan, or finance public transportation. An important element of the research was the development of a framework for transit agencies to use for prioritization of capital asset rehabilitation and replacement decisions. The framework builds upon a set of fundamental concepts and provides a basic set of steps for transit agencies to follow when evaluating and prioritizing rehabilitation and replacement investments. An analytical approach and set of spreadsheet tools were developed to support the framework. These address how to evaluate rehabilitation and replacement actions for specific types of transit assets and how to prioritize candidate rehabilitation and replacement actions. A detailed example is provided that demonstrates application of the analytical approach and tools in support of the framework. Practitioners, researchers, and transit agencies can use the results of the research to better prioritize their investments in existing capital assets and better communicate the predicted impacts of a given set of rehabilitation and replacement investments. The results of the research are intended to be of immediate value for transit agencies. In addition, several areas have been identified through this effort where additional research may be merited to support further improvements in assessing and addressing SGR concerns. These areas include the following: • Implementation guidance for the framework, analytical approach, and tools developed through this research effort;

B-2 Guidance for Calculating the Return on Investment in Transit State of Good Repair • Standards for asset data and condition assessment; • Synthesis of models and approaches for track and track-related assets used in passenger and freight rail in the United States and abroad; • Research on the relationship between asset condition and user impacts, such as delay; • Improved high-level models for relating investment levels to performance; • Quantification of transit agency prioritization strategies; and • Guidance on applying asset management concepts to transit. Spy Pond Partners, LLC; Harry Cohen, and Katherine O’Neil. TCRP Report 172: Guidance for Developing a Transit Asset Management Plan. Transportation Research Board of the National Academies, Washington, D.C., 2014. This document provides a process for developing a transit asset management plan used by transit agencies seeking to achieve SGR. The report is accompanied by a Transit Asset Prioritization Tool (TAPT), which is composed of four spreadsheet models designed to assist transit agencies in predicting the future conditions of their assets and in prioritizing asset rehabilitation and replacement. TCRP Report 172 describes a step-by-step process for developing a transit asset management plan, in accordance with MAP 21 requirements, and presents guidance on how to use TAPT to predict asset performance and prioritize rehabilitation and replacement actions. Spy Pond Partners, LLC; AECOM; McCollom Management Consulting, Inc.; Harry Cohen; Steven Silkunas; and David Hughes. TCRP Research Report 198: Relationship Between Transit Asset Condition and Service Quality. Transportation Research Board, Washington, D.C., 2018. This report presents a framework for relating service quality to asset condition. It describes different characteristics of service quality and discusses the relationship between these characteristics and asset condition. It describes a service quality measure, equivalent journey time as a single measure that captures time involved in different components of a passenger’s journey, with adjustments for the passenger’s perceptions of service. Three cases studies are detailed illustrating prediction of EJT and impact of asset condition on this measure. The report also describes the remaining tasks of TCRP E-11 to build on the framework and case studies and provide practical guidance for transit agency use. The report does not detail relationship of service quality to travel demand or broader social and economic impacts, which are the focus of this research. Travel Demand Literature Value of Reliability Arentze, T. A., and E. J. Molin. “Travelers’ Preferences in Multimodal Networks: Design and Results of a Comprehensive Series of Choice Experiments.” Transportation Research Part A: Policy and Practice, 58, 2013, pp. 15–28. The authors conduct a series of stated preference experiments to estimate route choice models in

Annotated Bibliography B-3 multimodal networks, including options such as combinations of car and public transportation. They specify reliability as a fixed 0.2 probability of a delay of x minutes occurring, where x is varied between 0, 10, 15, 20, or 30 minutes, depending on the experiment. The estimation results show the value of a minute of delay occurring with 0.2 probability to be equal to 0.68 minutes of in-vehicle travel time (IVTT) (or 3.4 minutes for a delay occurring with probability 1). However, a concern regarding this specification is the fixed value of probability (see Börjessen, 2011, for a discussion). Bates, J., J. Polak, P. Jones, and A. Cook. “The Valuation of Reliability for Personal Travel.” Transportation Research Part E: Logistics and Transportation Review, 37(2), 2001, pp.191–229. The authors present the three theoretical models for incorporating unreliability in travel decision making and the state of research at the time of writing. While progress has been made since then, this paper remains relevant in particular to public transportation researchers, given that much of the work so far has focused on automobile transportation. Bates et al. write that there is some evidence suggesting that, in public transportation, travelers may dislike failure to adhere to the schedule per se, in addition to the effect on travel times (no citation provided). The authors then present, in a comprehensive manner, the effects of transit unreliability on users’ travel time distributions. For a single segment, the travel time distribution is composed of the distribution of the wait time and the distribution of the in-vehicle travel time, which are likely not to be independent. This, in combination with the effect of headways and transfers, leads to travel time distributions that are generally complex and not well approximated by known distributions like the normal distribution. This has large implications in terms of the implied cost of unreliability, and to date, these problems have not been resolved. Based on published results at the time of writing, the authors suggest that reliability ratios around 1.3 appear plausible for car travel, and somewhat higher values for public transportation, but most likely not above 2. Börjesson, M., and J. Eliasson, J. “On the Use of ‘Average Delay’ as a Measure of Train Reliability.” Transportation Research Part A: Policy and Practice, 45(3), 2011, pp. 171–184. The authors argue that, in general, a reliability measure that is used for cost–benefit analysis should be consistent with the microeconomic framework in which travelers maximize the expected value of some utility function. Beyond that, the adequate choice of a measure of travel time variability is largely an empirical issue. For non-scheduled services with high travel time variability, the mean– variance approach has been shown to be consistent with underlying behavioral models under certain assumptions. On the other hand, for scheduled services with high reliability and long headways (such as intercity rail or commuter rail), “average delay per train” is a common reliability metric used in the UK, Sweden, and other European countries. With the help of three theoretical models of consumer behavior, they show that under some general assumptions, the disutility of delays will not increase linearly with the delay probability, with the average delay length, or with both. They then estimate three empirical models from stated preference surveys conducted in Sweden, and they are able to show that with those data, the value of

B-4 Guidance for Calculating the Return on Investment in Transit State of Good Repair delays consistently increases more slowly than linearly with the delay risk, and hence that the reliability multiplier (how many minutes of travel time a minute of average delay is worth) will in fact depend on the risk level—the lower the risk, the higher the reliability multiplier. However, with a fixed risk level, the value of delays seems to increase approximately linearly in the delay length in most cases. In other words, reducing the delay risk by half will reduce the value of delays by less than half, and doubling the delay risk will less than double the value of delays. This makes “average delay” a problematic reliability measure: it will tend to underestimate the value of eliminating small risks of long delays, and the reliability multiplier cannot necessarily be compared across studies since it will depend on the risk. Börjesson, M., J. Eliasson, and J. P. Franklin. “Valuations of Travel Time Variability in Scheduling Versus Mean–Variance Models.” Transportation Research Part B: Methodological, 46(7), 2012, pp. 855–873. The authors empirically estimate the VOR using the scheduling model and using a reduced-form specification, based on data from public transportation users in Sweden. According to recent theoretical findings, the model results should be comparable. However, the valuation of reliability implied by the scheduling model turns out to be substantially smaller than the results from the reduced-form model. Given these results, the authors question some of the assumptions that underlie the derivation of reduced-form models from scheduling models. From a behavioral perspective, the authors conclude that a late arrival per se is not the only cause of disutility to passengers and that the delay risk itself has a negative impact on them. In the words of the authors, it seems as if being ‘‘delayed’’ is considerably worse than just being ‘‘late.’’ Buris, M., C. Spiegelman, A. Abir, and S. Lee. Travelers’ Value of Time and Reliability as Measured on Katy Freeway. Texas A&M Transportation Institute Report PRC 15- 37 F. 2016. This report discusses basic concepts involved in calculating VOT and VOR. It then describes an effort to quantify these with transponder data for travelers using managed lanes on the Katy Freeway in Houston, Texas. The calculated values were found to be significantly lower than that expected based on state preference studies. VOT was found to range between $0 and $26 per hour while the VOR was found to range between $3 and $8 per hour. The report includes a detailed review of literature on VOT and VOR, as well as an overview of federal and Texas guidance on these topics. Carrion, C., and D. M. Levinson. “Uncovering the Influence of Commuters' Perception on the Reliability Ratio.” Working Paper, University of Minnesota. 2012. One of the factors that makes the valuation of reliability challenging is that distributions of travel times and probabilities of delay are subject to perception errors by travelers. The perceived travel time distributions can differ substantially from objectively measured distributions, and it is the former that drives traveler decision making. Thus, perception errors also affect the calculation of reliability ratios in RP surveys. The authors present a study in which they calculated travelers’ valuation of reliability and the reliability ratio from both perceived (self-reported) travel times and from travel times derived from GPS data. The experiment involved automobile users in the

Annotated Bibliography B-5 Minneapolis–St. Paul area. The results show that the reliability ratio calculated from perceived travel times was 1.3 using a mean-standard deviation model but only 0.42 when calculated from the GPS data. Carrion, C., and D. Levinson. “Value of Travel Time Reliability: A Review of Current Evidence.” Transportation Research Part A: Policy and Practice, 46(4), 2012, pp. 720–741. This review article sums up the evidence on the VOR and the reliability ratio until 2011. The authors note that the studies published to that point differ along a number of dimensions, including experimental design, the theoretical framework, the type of unreliability measure, and the data source. As a consequence, the VOR exhibits a significant variation across studies. In a conceptual discussion, the authors distinguish between predictable (e.g., peak-hour congestion) and unpredictable (e.g., incidents) travel time variability. They further note that the travel time distribution is subject to perception errors by travelers, which affects decision making, and that in the past researchers have struggled with presenting unreliability to study participants in a meaningful way. Although the reliability ratio is sometimes observed to be higher among non-commuters, this can be due to lower values of time for that group of travelers. In a meta-analysis, the authors show that estimated reliability ratios have varied between 0.1 and 3.29 (uncorrected for the measure of dispersion). The authors propose a methodology to convert the various measures of dispersion to a standard deviation and then perform a meta-regression to explain differences between the studies. Corrected reliability ratios are not presented, and the meta-regression yields inconclusive results. The paper cites two authors (Ghosh, 2001, and Yan, 2002) who found RP estimates to be approximately twice as large as SP estimates. Fosgerau, M., and A. Karlström. “The Value of Reliability.” Transportation Research Part B: Methodological, 44(1), 2010, pp. 38–49. Starting with the scheduling model for reliability valuation, the authors develop the consumer’s decision problem when congestion is not changing over time and show that the optimal departure time is independent of the travel time distribution. They then derive an expression containing the mean travel time and a measure of reliability and thus show that under certain assumptions the mean–variance model can be derived from the scheduling model. For a conventional choice of parameters, roughly based on empirical evidence, a traveler’s optimal probability of being late is 0.2. It is shown that with these parameters, a mean–variance formulation that is consistent with microeconomic theory involves the mean and the unweighted average of all the percentiles beyond the 80th. This measure is interpreted as the “mean lateness in standardized travel time”. This formulation is close to older studies (e.g., Brownstone and Small, 2005) that used the difference between the 50th and the 80th percentile, but the latter lack a behavioral underpinning. See Small (2012) for a further discussion. Koster, P., E. Kroes, and E. Verhoef. “Travel Time Variability and Airport Accessibility.” Transportation Research Part B: Methodological, 45(10), 2011, pp. 1545–1559. The authors estimate a scheduling model for access to the airport by car. The goal of the

B-6 Guidance for Calculating the Return on Investment in Transit State of Good Repair paper is to estimate the “scheduling cost”—i.e., the additional cost of travel time to travelers because they need to leave their home earlier due to unreliability in travel times. The airport access cost function contains an additional cost term at the point where a traveler misses a flight. In a case study for access to Amsterdam Schipol Airport, the authors find that, on average, the scheduling cost is between 5 percent and 15 percent of total expected travel cost, but up to 30 percent in peak hours. It is close to 0 percent during the night. Costs for business travelers are slightly higher. Kouwenhoven, M., G. C. de Jong, P. Koster, V. A. van den Berg, E. T. Verhoef, J. Bates, and P. M. Warffemius. “New Values of Time and Reliability in Passenger Transport in The Netherlands.” Research in Transportation Economics, 47, 2014, pp. 37–49. The authors establish new values of time and reliability for the Netherlands as part of a national study. They note that the VOR had previously not been estimated in any national studies intended to produce values for policy analysis. Data were collected in 2009 and 2011, and it is observed that data collected via Internet surveys produced substantially lower values of time than traditional surveys. The authors test several model specifications and concluded that the mean-dispersion model performs well and that, all things considered, the standard deviation of the travel time distribution is the only realistically feasible measure for large-scale project evaluations. Their final latent class model was specified such that the VOR was estimated directly and was not class-specific, and it was estimated from a combined data set for automobile, intercity train, and metro/bus travel. The recommended reliability ratio is 0.4 for commute trips, 1.1 for business trips, and 0.6 for other trips. This works out to a VOR of EUR 3.25 for commute trips, 3.75 for other trips, and 21.75 for business trips. The VOR for business trips, which is considerably larger than the other values, is further decomposed into an employee part (EUR 12.00) and an employer part (EUR 9.75). The reliability ratio is compared to results from Norway (Ramjerdi and Flügel, 2010), which were found to be 0.69 for short trips and 0.42 for long trips. Li, Z., D. A. Hensher, and J. M. Rose. “Willingness to Pay for Travel Time Reliability in Passenger Transport: A Review and Some New Empirical Evidence.” Transportation Research Part E: Logistics and Transportation Review, 46(3), 2010, pp. 384–403. The authors present and review previous evidence on the VOR from Australia, Europe, and the United States, and note the high variability of the ratio across the studies they reviewed, between 0.1 and 3.3. The authors present the experiments by Small et al. using State Route 91 and Interstate 15 in Los Angeles as good benchmarks due to the incorporation of RP data and a very realistic experimental setup (a dynamic tolling scheme). They conclude that the way of presenting time variability may be a key contributor to different estimates, citing especially differences between work by the same researchers. The experiments in California were only focused on automobile travel. Regarding public transportation, the authors point to a study by Batley and Ibáñez (2009) on rail commuters in the UK, in which the reliability ratio was estimated at 2.1, and argue that public transportation is expected to exhibit a higher reliability ratio due to the more serious consequences of unreliability (missed connections, wait time for the next vehicle). This is especially

Annotated Bibliography B-7 true in large networks with many transfer passengers. Using a stated preference experiment with Australian automobile users and a mixed multinomial logit model, the authors then present new evidence. The reliability ratio is calculated to be 1.43 for commute trips and 1.78 for non-commute trips. The higher value of the latter is due to a lower VOT for non-commute trips. It is also shown that travelers with flexible arrival times placed a lower value on reliability (by almost 50 percent) than travelers with inflexible arrival times—an observation that was also made by Bhat and Sardesai (2006). Small, K. A. “Valuation of Travel Time.” Economics of Transportation, 1(1), 2012, pp. 2– 14. Small sums up current evidence on the value of travel time, including the VOR. The value of travel time for commute trips seems to hover around one half the gross wage rate, and while it is indisputable that there is a cost to unreliability, empirical evidence on the value of unreliability is less conclusive. The impact of unreliability consists of two costs to the traveler: The direct cost of delays and the routine costs of departing earlier to allow a safety margin. Based on Fosgerau and Karlström (2010), Small presents two formulations for a reliability measure: one that includes the “mean lateness in standardized travel time” in the VOR (the conventional approach) and one that does not. He shows that using the conventional approach make the estimated VOR dependent on the shape of the travel time distribution, which limits transferability of results. Wardman, M., V. P. K. Chintakayala, and G. de Jong. “Values of Travel Time in Europe: Review and Meta-Analysis.” Transportation Research Part A: Policy and Practice, 94, 2016, pp. 93–111. In a large meta-analysis of monetary valuations of travel time (and reliability) across Europe, the authors compile evidence from 389 studies across all surface modes and covering both urban and inter-urban travel. They find RP valuations of time to be significantly larger than stated preference valuations, by an average factor of 1.46 for business and 1.38 for non-business trips. Valuation of commuting time is, on average, 1.13 times higher than non-commuting time for train travel, and 1.37 times for bus travel. Valuation of metro travel time is found to be around 0.8 times the valuation of bus travel time. A minute of headway is found to be valued at 0.75 minutes of in-vehicle time. Reliability is included in the meta-model in four different terms, reflecting the different ways in which it was specified by researchers. The model shows a valuation of a minute of late time at 3.1 times IVTT, of schedule delay late/early at 1.97/0.92 minutes of IVTT, and 1 minute of standard deviation at 0.64 minutes of IVTT, across all modes. The authors note that there is a wide range in study and data quality. Elasticities Barron, A., P. Melo, J. Cohen, and R. Anderson., “Passenger-Focused Management Approach to Measurement of Train Delay Impacts.” Transportation Research Record: Journal of the Transportation Research Board, No. 2351, 2013, pp. 46–53. The authors conduct a benchmarking survey of 22 metro operators (names withheld) that are part of the CoMET or NOVA industry associations. They request detailed information on the incident tracking policies in place, including, among other

B-8 Guidance for Calculating the Return on Investment in Transit State of Good Repair elements, incident causes, operational impacts, and passenger impacts. They find that 21 systems were able to provide basic incident data, but that only two systems tracked the total number of passengers affected, and only one system calculated total passenger delay. Both systems are located in Asia. Carrel, A., A. Halvorsen, and J. Walker. “Passengers' Perception of and Behavioral Adaptation to Unreliability in Public Transportation.” Transportation Research Record: Journal of the Transportation Research Board, No. 2351, 2013, pp. 153– 162. In this paper, the results of a survey among transit users and former transit users in San Francisco are presented. The survey first asked respondents to rate the importance of various reliability aspects. Then, a list of different types of unreliability incidents was presented, and respondents were asked to indicate when they had last experienced them and what strategies they employed to adapt to unreliability on the network. Reliability was valued higher for commute trips than for non-commute trips, and reliability of transfers proved to be the most important characteristic. The ability to find a seat was seen as less important, but the model (presented below) showed that not being able to board a vehicle due to crowding had a negative influence. In line with previous research, it was found that on average, respondents considered a 10-minute headway to be “frequent service.” A large portion of users reported actively adapting their behavior to deal with unreliability, and respondents reported both switching to other transit modes that were perceived to be more reliable (e.g., local bus to rail), and switching to non- transit modes. Although the sample size of former riders was very small, 50 percent of them indicated unreliability as a reason for transit use cessation. The authors estimated an ordinal logit model to explain reduction or cessation of transit use as a function of previous experiences with unreliability. Demographic information, bike ownership, and automobile ownership were found to be insignificant (which was not surprising since the surveyed population consisted mostly of choice riders in San Francisco). In general, it appeared that passengers were more forgiving of problems that were perceived to be outside of the transit agency’s control, and the strongest influence on transit use reduction came from experiencing on-board delays due to operational problems. Furthermore, it was found that the point in a person’s trip at which the delay was encountered mattered, with delays at transfer stops and delays while on board having a stronger influence than delays at the origin stop. The paper also presents some findings regarding the use of real-time information: Few passengers reported knowing the timetable, and the authors hypothesize that the use of real-time information may contribute to the perception of unreliability. Frei, C., and H. Mahmassani. “Riding More Frequently: Estimating Disaggregate Ridership Elasticity for a Large Urban Bus Transit Network.” Transportation Research Record: Journal of the Transportation Research Board, No. 2350, 2013, pp. 65–71. In this paper, stop-level transit elasticities with respect to service frequency are estimated. The focus on stop-level boardings helps avoid the inflation of ridership numbers that can be caused by aggregating route or network ridership. The headway (arc) elasticity of ridership is found to be around −0.27, very similar to an earlier disaggregate analysis of New York City data.

Annotated Bibliography B-9 Kroes, E., P. Koster, and S. Peer. “A Practical Method to Estimate the Benefits of Improved Road Network Reliability: An Application to Departing Air Passengers.” Transportation, 45(5). 2018. The authors present a methodology to estimate the impact of changes in travel time variability for car travel. For every OD pair, the mean delay must be known, and the standard deviation of the delay is then estimated as a constant factor (suggested to be approximately 0.8) times the mean delay. This result is based on an earlier paper by the authors. The approach is implemented in a large-scale road network model, and it is shown that for every 1- Euro reduction in mean travel time costs due to delays, there is an additional cost reduction of about 0.75–0.85 Euros due to lower travel time variability and hence scheduling costs. Litman, T. Understanding Transport Demands and Elasticities: How Prices and Other Factors Affect Travel Behavior. Victoria Transport Policy Institute. 2013. In this guide, Litman conducts a broad review of published elasticity values for public transportation. Generally speaking, there are no published elasticities with respect to reliability, but elasticities with respect to travel times or generalized costs could be used as an approximation. The author argues that the elasticity of transit demand depends heavily on the number of discretionary riders, and that increased travel times lead to the loss of travelers with high values of time first. There is some evidence that demand elasticities with respect to transit fares may be non-symmetric (i.e., that losses are greater than gains), since ridership losses can lead to increases in auto ownership. If this is also the case with travel times and reliability, it would mean that it is difficult to attract users back to transit after improvements are made to achieve SGR. Long-run elasticities over 1–3 years are generally larger than short- run elasticities, by a factor of 2–3. This is again due to the fact that, with time, consumers have more opportunities to adapt and make vehicle ownership and residential location decisions. Consumers tend to be more responsive to changes they consider durable. Increased travel speeds and reduced delays across all modes tend to increase travel distances (in line with the theory of constant travel time budgets), whereas increased relative speed for a particular mode tends to attract travel from other modes on a corridor. With regard to travel time elasticities, off- peak travel tends to be more elastic by a factor of 1.5–2. A potential difficulty in using elasticities is that the impact of not maintaining a transit system is gradual; published elasticities, on the other hand, typically relate to instantaneous changes, such a fare increase or timetable change. Preston, J., G. Wall, R. Batley, J. Ibáñez, and J. Shires. “Impact of Delays on Passenger Train Services: Evidence from Great Britain.” Transportation Research Record: Journal of the Transportation Research Board, No. 2117, 2009, pp. 14–23. This paper summarizes prior research investigating the effect of delay times on rail demand in the UK. The official Passenger Demand Forecasting Handbook used by UK train operating companies recommends a reliability multiplier of 3 and a reliability ratio of 1. The authors cite an early RP study (Rail OR, 1996) that showed a lag of approximately 6 months before performance affected revenue. The authors present results of two studies (conducted in 2003, 2005) in which elasticities of demand with

B-10 Guidance for Calculating the Return on Investment in Transit State of Good Repair respect to delay are derived. They are surprisingly low, especially for the London and Southwest England region, where commuting dominates, auto ownership is low, and many users are captive to rail. The authors review recent work on the “value of delay time” and conclude that, while it is clearly important, study results tend to vary and are difficult to generalize. Wardman, M. “Review and Meta-Analysis of UK Time Elasticities of Travel Demand.” Transportation, 39(3), 2012, pp. 465–490. The author conducts a broad analysis of 427 published demand elasticities with respect to travel time in the UK. Overall, he notes that quite strong effects of distance have been detected, but that there is little apparent variation by trip purpose, type of data (e.g., revealed preference or stated preference), or region. On the other hand, he points out that there tend to be large variations in elasticities across countries, so it is unclear how transferable the UK evidence is. In the UK, generalized journey time (GJT) is a frequently used measure, whereby the different trip stages are converted into equivalent time units. Overall, the average elasticities tend to be <1 in the short run, but long-run elasticities are, on average, between 2 and 3 times as large as short-run elasticities and may well be >1. The “long run” is defined differently by different studies, but it typically is between 3 and 5 years. The meta-analysis shows that commuters are slightly more sensitive to travel times, though not at a statistically significant level. Based on an unpublished study by Steer Davies Gleave (2004) that looked at car demand elasticities, the author notes that congestion is typically not considered in elasticity studies, though this might be an important variable. Effects of Service Quality Peer, S., E. Verhoef, J. Knockaert, P. Koster, and Y. Y. Tseng. “Long-Run Versus Short-Run Perspectives on Consumer Scheduling: Evidence from a Revealed- Preference Experiment Among Peak-Hour Road Commuters.” International Economic Review, 56(1), 2015, pp. 303–323. This paper takes an in-depth look at the scheduling behavior of Dutch travelers, and the results reinforce the notion that short-run and long-run valuations of travel times are different, mainly because in the long run, travelers have the opportunity to reschedule activities and departure times. Schedule deviations (schedule delay late/schedule delay early), on the other hand, are found to be valued more highly in the short run, since there are typically more binding constraints at that point. Two models are presented, a short-run and a long- run model, but travel time variability is not included because in practice it is often highly correlated with the mean travel time (Note: this is true of revealed preference data, not stated preference). The long-run VOT is found to be higher than the short- run VOT, which may partly be due to an implicit inclusion of variability in that term; with the help of theoretical considerations, the authors conclude that about half of the long-run VOT they found can be attributed to travel time variability. Van Oort, N., T. Brands, E. de Romph, J. A. Flores, and B. Eng. “Unreliability Effects in Public Transport Modelling.” International Journal of Transportation, 3(1), 2015, pp. 113–130. The authors propose a three-step approach to including unreliability in a

Annotated Bibliography B-11 network flow model for public transportation networks. A main goal of this study was to include unreliability without changing the modeling software requirements from what is currently used. They discuss the various components of transit travel time unreliability and point out that, aside from travel times, unreliability also affects public perception and crowding levels. A generalized cost function is defined, in which they include an overall travel time consisting of wait time, in-vehicle time, additional wait time due to unreliability, and the variances of both in-vehicle time and wait time. It is shown how all components can be derived from automatic vehicle location data. Two strategies are discussed: incorporating the additional travel times at the stop level and at the line level. The model parameters are not estimated, but rather selected from previous literature. In a case study for the city of Utrecht, the researchers show that including reliability at a stop level produces, on average, a more accurate representation of network flows. Including reliability at the line level does not produce a significant improvement. Van Oort, N. “Incorporating Enhanced Service Reliability of Public Transport in Cost– Benefit Analyses.” Public Transport, 8(1), 2016, pp. 143–160. The author discusses and applies the framework presented previously by Van Oort et al. He points out that the London Underground is one of the few transit authorities to track passenger impacts of delays in the form of a metric called “Excess Journey Time.” In the case study, the author uses two measures of unreliability, additional travel time (especially wait time) and the standard deviation of travel time. The former is multiplied by a “standard” VOT, and the latter is multiplied by the VOR (i.e., value of standard deviation) reported by RAND and AVV in 2005. The VOR makes up 2/3 of the benefits of the project being discussed. The author notes that there is not yet a methodology to quantify the benefit of reliability on multi-leg trips. Vovsha, P., G. S. O. Marcelo, and D. William. “Statistical Analysis of Transit User Preferences Including In-Vehicle Crowding and Service Reliability.” Transportation Research Board 2014 Annual Meeting Compendium of Papers. 2014. The authors describe the development and analysis of a large-scale stated preference survey in Los Angeles. The survey data was used to model the effects of crowding and on- time performance on transit demand. After a literature review, schedule adherence at boarding stop (expressed as average extra wait time at boarding, with an associated frequency) was chosen as a reliability indicator for practical reasons. The results show that the effect of crowding, expressed in terms of a travel time weight, ranges from 1 (no crowding) to above 1.6 (high crowding). Some variations were found as a function of socio-demographic characteristics, trip type, and transit mode. Regarding delay probability, the authors found a certain convexity with respect to delay times (longer delays cause larger disutility) and a concavity with respect to probability of delay: delays under 3 minutes seem to be largely disregarded by passengers, and a behavioral adaptation (shift in departure times) appears to occur around delay probabilities of 0.5 to 0.6. The model was incorporated in the operational model used by LACMTA.

B-12 Guidance for Calculating the Return on Investment in Transit State of Good Repair Predicting Ridership Chakrabarti, S. “The Demand for Reliable Transit Service: New Evidence Using Stop- Level Data from the Los Angeles Metro Bus System.” Journal of Transport Geography, 48, 2015, pp. 154–164. The author estimates a negative binomial regression model of stop-level boardings for Los Angeles MTA buses. The explanatory variables include the headway and two reliability measures: average on- time performance (OTP), defined as the average percentage of departures between 1 minute early and 5 minutes late at a given stop, and the standard deviation of the schedule deviation. The results show that, consistent with literature, service frequency is a major determinant of ridership, and both OTP and standard deviation are found to have an effect, though the effect of increased standard deviation is, counterintuitively, found to be negative. Furthermore, an effect of OTP is only observed during peak hours, and the implied elasticity for OTP is relatively small. Iseki, H., and R. Ali. “Fixed-Effects Panel Data Analysis of Gasoline Prices, Fare, Service Supply, and Service Frequency on Transit Ridership in 10 U.S. Urbanized Areas.” Transportation Research Record of the Transportation Research Board, No. 2537, 2015, pp. 71–80. This paper estimates a log-log regression model to predict transit ridership. The model specification allows elasticities to be estimated directly. The authors note that, in general, omitted variable bias is relatively prevalent and difficult to account for in ridership prediction studies. They cite Taylor et al. (2009) as including the most comprehensive list of influential factors in their cross-sectional regression analysis to predict ridership. A further bias they warn of is endogeneity that arises from the bidirectional causality between transit supply and demand. They then present a fixed-effects panel data model to predict monthly unlinked passenger trips in ten urban areas in the U.S. Two of the input variables relate to transit SGR: service frequency and vehicle revenue hours. Estimated elasticities for overall demand and demand by mode are reported for two different model specifications, but some of the estimates are insignificant, which the authors attribute to a lack of variance in the data. They further observe that half of the urban areas included in the study have a higher proportion of zero-vehicle households than the national average, and that zero-vehicle households tend to be less price-sensitive than discretionary riders and, at least in the short run, also can be less sensitive to changes in-service supply. Paterson, L., and D. Vautin. “Evaluating the Regional Benefit/Cost Ratio for Transit State of Good Repair Investments.” Journal of Public Transportation, 18(3), 2015. This paper examines the benefit–cost ratio of transit SGR investments and compares them to system expansion efforts. Using the TERM Lite model, TCRP Vehicle Model (from TCRP Report 157), a new model to quantify delay associated with the age of non-vehicle assets (such as fixed guideway, train control systems, etc.), the authors determine the impact of reduced SGR investments on operations. Then, with the help of a regional travel demand model, they link the operational outcomes with impacts on ridership. The methodology focuses on delays as the “primary operational impact of transit asset failure,” and direct costs are disregarded

Annotated Bibliography B-13 since they are not directly related to SGR maintenance funding. The authors investigate several funding scenarios, resulting travel time impacts, and regional benefits of SGR funding for 25 transit systems in the San Francisco Bay Area. The scenarios include a zero-funding option where the assets are allowed to degrade without any SGR investment, and a zero-regional-funding option where only the regional funding to SGR is eliminated. The authors suggest that there exists a point where increasing SGR funding is no longer economically efficient, but given the limitations of the modeling approach, they conclude that more scenarios would need to be analyzed to determine this point. Other References Li, Z. and D. Hensher. “Crowding in Public Transport: A Review of Objective and Subjective Measures.” Journal of Public Transportation, Vol. 16, No. 2, 2013. This paper has reviewed the specifications of crowding measures defined by transport authorities in different countries (UK, U.S., and Australia). The bus industry, including bus rapid transit (BRT), tends to use a generic measure, namely the number of standing passengers per square meter, while for the rail industry there are some variations in crowding measures (e.g., the number of standing passengers per square meter, load factor, rolling hour average loads). The authors suggest that for short journeys (e.g., commuting services), standing allowance should be treated as an additional component of capacity when defining crowding measures, while for long journeys (e.g., regional services), only the number of seats should be used as the capacity. The broad transport crowding literature tends to focus on objective measures (e.g., passenger density). Only a few transport studies (see, e.g., Turner et al. 2004; Cox et al. 2006; Mohd Mahudin et al. 2012) have argued that the objective treatment of crowding (equivalent to density) cannot fully represent the experience of crowding, given that the perception of crowding is subjective. Given this, in addition to the objective measures (e.g., density), public transport operators/authorities should conduct perception surveys to obtain information on passenger subjective evaluations of crowding. Through surveys on perceived crowding, the transport authorities/operators can obtain the real experiences of passengers, which can be used to design more appealing measures to capture crowding and to calibrate the defined crowding thresholds to reflect the experienced crowding. Incorporating subjective measures of crowding can contribute to (1) a more accurate representation of crowding, which would help operators manage and reduce crowding in time by implementing strategies such as increasing the frequency of service and using larger vehicles, and (2) a better understanding of crowding, which is beneficial to the design of more appealing public transport systems to attract more users. This evidence can be used on an ongoing basis to ensure that proxy measures of perceived preferences for specific levels of crowding that are incorporated in formal modal choice models that deliver the necessary outputs for benefit–cost analysis remain relevant. The challenge is to establish how much users are willing to pay to

B-14 Guidance for Calculating the Return on Investment in Transit State of Good Repair United States Government Accountability Office. Transit Asset Management: Additional Research on Capital Investment Effects Could Help Transit Agencies Optimize Funding. 2013. Utilizing proper tools to assess the condition and performance of transit assets and determining the effects of capital investments on system performance would allow transit agencies to optimize their capital improvement plans in light of limited funding. As both ridership and funding constraints rise for the nation’s public transportation systems, prioritization of the capital improvements backlog becomes more important in order to maximize the investments’ impact on system performance. The study presented in this paper examined the practices used by agencies to measure the impact of capital investments, highlighted FTA initiatives to support transit agencies asset management practices, and reviewed case studies that examine the effect of capital investment decisions on system performance and ridership in Chicago and New York City. A case study that reviewed the effects of $74 billion worth of capital improvement projects spanning from the 1980s to 2008 on the performance of New York City’s transit system found that total system ridership increased by 58 percent from 1982 to 2007. This compares with a 17 percent drop in ridership that was observed in the 1970s. It is important to note that there are many factors that may have jointly contributed to the increase in ridership. Some of these factors include recovery of the City of New York’s financial situation, reduction in crime, and growth of financial and service sectors. The report does not break out what portion of the effects can be attributed to capital improvement or other details of the case study. While more studies need to be conducted on the effect of capital investments on ridership, this shows a positive relationship between capital investment and ridership. Economic and Social Impacts Literature Economic Impacts of State of Good Repair Investments Amber Ontiveros Associates, LLC, and Econsult Solutions. State of Good Repair and Transit Equity. 2015. “Transportation equity centers on the notion that low-income and minority communities should have equal access to high-quality transit services, and that the benefits and burdens of transportation projects should be distributed equitably.” The analysis presented in this paper focuses on the implications of transit equity in decision making for transit service capital investments. The paper demonstrates that the principles and methodologies that underlie sound equity analyses of proposed transit service changes can be applied to evaluate SGR reduce crowding (to a specific level), as they perceive it, regardless of the standard, since this is a clear source of user benefit. Mapping this evidence, if available, to the standard, will enable a clearer picture to emerge of how the system is complying with the standard; however, this is not the basis of extracting the set of crowding- related benefits that exist regardless of the standard. If a move toward the standard ensures a gain in perceived user benefit, then it needs to be captured through a preference study. Simply imposing a desired standard does not capture the user benefit.

Annotated Bibliography B-15 expenditures. Furthermore, a new methodology—“ridership-adjusted population”—is outlined for conducting service equity analysis where rider’s home location/origin data is available. This approach improves the accuracy of the analysis and validity of findings. Using data from Southeast Pennsylvania Transit Authority (SEPTA), Section 4 of the paper illustrates the application of the ridership-adjusted population methodology to a service equity analysis, at both system-wide level and specific route level. Analysis showed that the ridership-adjusted population profile of the SEPTA service area is found to be 49.6 percent minority and 32.1 percent low-income households. In comparison, the known minority and low-income proportions of the service area based on American Community Survey (ACS) data is 36.6 percent and 21.7 percent, respectively. At a route-specific level, the ridership-adjusted population profile of the Market-Frankford Line is found to be 53.2 percent minority and 35.6 percent low- income households. System-wide analysis demonstrated that SEPTA’s existing ridership base disproportionately serves minorities and low-income households relative to the population of the service area. The route-specific analysis demonstrates that the Market-Frankford Line serves an even greater concentration of minorities and low- income households than the system as a whole. The ridership-adjusted population methodology for service equity analysis helps identify the general appeal of SGR programs to maintain existing transit services from an equity perspective. Additionally, by providing an understanding of the equity implications of investments into existing and new routes, the ridership-adjusted population methodology can also be used to allocate resources between existing and new services. ECONorthwest and Parsons Brinckerhoff Quade and Douglas, Inc. TCRP Report 78: Estimating the Benefits and Costs of Public Transit Projects: A Guidebook for Practitioners. TRB, National Research Council, Washington, D.C., 2002. A Guidebook for Practitioners provides a resource for people who have the difficult and often cumbersome responsibility of analyzing the benefits and costs of public transportation services and presenting the results of these analyses to decision makers, the media, and the public. The researchers prepared a guidebook and CD-ROM (CRP-CD-18), which constitute TCRP Report 78. The research results are of particular interest to individuals who plan and evaluate the benefits and costs of new investments in public transportation. Other audiences for this report include policymakers, transportation activists, other transportation professionals, and students in related fields. This report synthesizes both theory and empirical work to provide practical methods for estimating the benefits and costs of many typical transit improvements. It is written primarily for transit planners in state, regional, and local government responsible for evaluating transit investments.

B-16 Guidance for Calculating the Return on Investment in Transit State of Good Repair Litman, T. Evaluating Public Transit Benefits and Costs: Best Practices Guidebook. Victoria Transport Policy Institute. 2017. This guidebook describes how to create a comprehensive framework for evaluating the full impacts (benefits and costs) of a particular transit service or improvement. It identifies various categories of impacts and how to measure them. It discusses best practices for transit evaluation and identifies common errors that distort results. It discusses the travel impacts of various types of transit system changes and incentives. It describes ways to optimize transit benefits by increasing system efficiency, increasing ridership, and creating more transit-oriented land-use patterns. It compares automobile and transit costs and the advantages and disadvantages of bus and rail transit. It includes examples of transit evaluation and provides extensive references. Many of the techniques in this guide can be used to evaluate other modes, such as ridesharing, cycling, and walking Lu, Lexcie. The Relationship Between Maintenance and Service Planning, and Evaluation of Maintenance Strategies (or Should You Run Your EMUs Till They Drop?). 2003. Carrying out maintenance activities during operating hours could lead to greater reliability and lower maintenance costs. However, the associated reduction in transit service levels could result in passenger delays and customer inconvenience. Traditionally, the trade-off between transit service reliability and service frequency is not thoroughly evaluated in transit maintenance and service planning. To assess system impacts, an approach that considers maintenance costs, service reliability, and customer impacts due to reduced transit service within a single framework is needed. This paper presents a theoretical framework and an example methodology within transit maintenance and service planning to compare the benefits of increased transit maintenance with customer impacts such as overcrowding and travel delays. To demonstrate the implementation of the revised framework, a decision support model that is sensitive to the Massachusetts Bay Transportation Authority’s (MBTA’s) operating plan was developed and tested for MBTA’s Orange Line. An operating plan was proposed for MBTA’s Orange Line, which would decrease the peak fleet requirement from 102/120 cars to 84/120 cars. The proposition entails decreasing the rush-hour headway from every five minutes to four minutes, but running four-car consists instead of six-car consists. To quantify the effect of possible overcrowding versus the benefit of increased maintenance, a revised framework/model is developed. Economic evaluations within the revised framework include • Economic cost of passengers beyond planned capacity; • Cost of providing additional operators for the consists; • Cost of unreliability, which depends on the probability of failure and the impact of failures on customers; • Cost of waiting time, which depends on the headway; and

Annotated Bibliography B-17 • Expected life-cycle cost of the equipment, based on the operations and maintenance plan. Results from the MBTA Orange Line study showed that, while increased operator costs associated with increased frequency are significant, the benefits associated with reduced failures (due to increased maintenance activity/reduced fleet requirements) dominate operator costs. However, in a capacity-constrained situation, the costs of passenger spills (when transit demand exceeds capacity) dominate both. The value of reduced wait time for existing riders dominates all costs. It is important to note that the model underestimates costs associated with passenger spills as it does not account for cascading delays in the transit system. Calculating Return on Investment for State of Good Repair American Society of Civil Engineers. Failure to Act: The Economic Impact of Current Investment Trends in Surface Transportation Infrastructure. 2011. The American Society of Civil Engineers (ASCE) report analyzes the economic implications of surface transportation infrastructure deficiencies (highways, bridges, rail and transit) on the U.S. economy. In 2010, it was estimated that deficiencies in America’s surface transportation systems cost households and businesses nearly $130 billion. Costs include excess vehicle operating costs, travel time delays, safety, and environmental costs. The following table summarizes the cost of deficient and deteriorating conditions in pavement, bridges, transit vehicles, track, and transit facilities in year 2010 and how these costs are expected to accumulate by 2040. Table 1. ASCE Report, Failure to Act: The Economic Impact of Current Investment Trends in Surface Transportation. Performance Area Cost of Deficiencies (billions of constant 2010 dollars) In 2010 By 2020 By 2040 Pavement and Bridge Conditions $10 $58 $651 Highway Congestion $27 $276 $1,272 Rail Transit* $41 $171 $370 Bus Transit* $49 $398 $659 Inter-City Rail* $2 $10 $20 Total Cost to System Users $130 $912 $2,972 *deficiencies in transit vehicle, track, and transit facilities In the case of transit, deficiencies may compound among modes. For example, fixed-route transit bus or demand-response vehicles that are in poor condition may also operate in congested conditions on deficient pavements. Based on EDR Group’s synthesis of results from the USDOT Transit Economic Requirements Model (TERM) model and 2010 National Transit Database (NTD), the

B-18 Guidance for Calculating the Return on Investment in Transit State of Good Repair United States carries a backlog of $86 billion in unfunded transit capital investment needs. In 2010, approximately 15 percent of transit revenue miles were on vehicles with an SGR of “fair” or “poor.” Based on report findings, assuming current funding levels, • 16.2 percent of transit bus VMT in 2010 occurred in suboptimal conditions; this is expected to increase to 30 percent in 2040; • 16 percent of demand-response bus VMT in 2010 were on deficient vehicles; this is expected to increase to 68 percent in 2040; • 7 percent of light rail VMT in 2010 were on deficient vehicles; this is expected to increase to 22 percent in 2040; and • 11.2 percent of other rail vehicles in 2010 were deficient; this is expected to increase to 15.8 percent in 2040. Between 2010 and 2040, the U.S. population is expected to grow by one-third, and the proportion of Americans aged 75 years and older is expected to nearly double. Population depending on demand-response transit systems would increase tremendously. The study points to a need for more research into adequacy of today’s demand-response fleets for a growing non-driving population. Cambridge Systematics, Inc., with Vlecides-Shroeder Associates, Inc., B. Welsh, and Ernest Sawyer Enterprises, Inc. Investment in Public Transportation: The Economic Impacts of the RTA System on Regional and State Economies. 1995. The study described in this document evaluates the overall economic benefits and costs of RTA’s capital investment programs by comparing the benefits and costs with the state and regional economies associated with several different levels of capital investment by the RTA. Several scenarios were developed to evaluate their effects over a 20- year timeframe (1995–2014): • Baseline/deterioration scenario, which maintains the 1994 RTA funding levels, the time during which the study was conducted; • Disinvestment scenario, which reduces funding from the baseline to a minimum needed to keep the system operating; • SGR scenario, which brings the funding to a level that is required to keep the system in a state of good repair; and • System Expansion, which would expand service and upgrade the quality of vehicles and facilities. The study considered a number of benefits, including those incurred by travelers on both the transit system and highways, and assessed how the changes would affect the cost of doing business and spending patterns. Costs and benefits were discounted using 4 percent and 10 percent discount rates. The results of the study showed that bringing the system up to SGR yielded over $6 in benefits for every dollar invested, both at 4 percent and 10 percent discount rates. The System

Annotated Bibliography B-19 Expansion scenario, on the other hand, yielded $1.8 and $1.5 in benefits for every dollar invested, at 4 percent and 10 percent discount rates, respectively. Forkenbrock, D., and G. Weisbrod. NCHRP Report 456: Guidebook for Assessing the Social and Economic Effects of Transportation Projects. TRB, National Research Council, Washington, D.C., 2001. This report presents guidance for practitioners in assessing the social and economic implications of transportation projects for their surrounding communities. Presented in guidebook format, the report identifies current best methods, tools, and techniques based on an extensive literature review and comprehensive survey of state departments of transportation and metropolitan planning organizations. Additional resources are contained in the appendices, including a discussion of geographic information system applications for social and economic impact analysis, tips on designing effective survey questionnaires, an overview of the travel demand modeling process, and a brief review of relevant legislation that provides the legal basis for impact assessment requirements. Eddington, R. The Eddington Transport Study: The Case for Action: Sir Rod Eddington’s Advice to Government. 2006. This study demonstrates that the performance of the UK’s transport networks will be a crucial enabler of sustained productivity and competitiveness: a 5 percent reduction in travel time for all business travel on the roads could generate around £2.5 billion of cost savings—some 0.2 percent of GDP. Good transport systems support the productivity of urban areas, supporting deep and productive labor markets and allowing businesses to reap the benefits of agglomeration. Transport corridors are the arteries of domestic and international trade, boosting the competitiveness of the UK economy. Correspondingly, transport policies offer some remarkable economic returns with many schemes offering benefits several times their costs, even once environmental costs have been factored in. Schwieterman, J., Audenaerd, L., and Schulz, M. Tending to Transit: The Benefits and Costs of Bringing Public Transport in the Chicago Region into Good Repair. 2012. The study described in this paper examines the costs and benefits of bringing the transit system in Chicago to SGR. The study concluded, among other things, that making just a minimal level of investment into the system would detrimentally affect system ridership. The report cites several studies that indicate that such an approach to capital improvements can contribute to a ridership drop of between 15 and 20 percent. The report also highlighted a 2007 study titled “Chicago Metropolis 2020” that estimated that $1 of investment into the transit system would return between $1.21 and $1.64 in benefits. Other studies for the Chicago area put the return on benefits as high as $1.90 for every $1 invested, and studies in other regions have shown returns as high as $3.41. Additional benefits, like the system’s impact on tourism and land-use impacts, were not considered. Chicago Metropolis 2020. Time Is Money: the Economic Benefits of Transit Investment. 2007. The study described in this paper estimates the expected return on transit investments into Chicago’s transit system in accordance with Regional

B-20 Guidance for Calculating the Return on Investment in Transit State of Good Repair Transportation Authority’s (RTA’s) Strategic Plan, “Moving Beyond Congestion.” The analysis used conservative assumptions to model the changes in the region’s travel patterns and calculate the cost of time delays, accidents, air pollution from congestion, and direct cost to drivers. Several scenarios were evaluated: • Decline, which assumes that there will be no new funding for operations or capital improvements for transit, leading to a slow decline in service; • Maintain, which will maintain the existing service levels; • Expand, which will go beyond simply maintaining the level of service of the system and will add new service and resolve system issues, as recommended in RTA’s Strategic Plan; and • Expand and Plan, which will make the same assumptions as the Expand scenario but will assume greater emphasis on transit-oriented development by using land-use concepts from CMAP’s 2040 “Regional Framework Plan.” The analysis estimated that maintaining the system would generate a return of $1.21 for every dollar invested, relative to the Decline scenario. Under the Expand scenario the investment would generate a return of $1.34 for every dollar invested. Under the Expand and Plan scenario, $1.61 in benefits is estimated to be generated for every dollar invested. American Public Transport Association (APTA). Economic Impact of Public Transportation: 2014 Update. 2014. This report focuses on how investment in public transportation affects the economy in terms of employment wages and business income, updating a previous report prepared for TCRP Project J-11, Task 7 and described in the main body of this report. It specifically addresses the issue of how various aspects of the economy are affected by decisions made regarding investment in public transportation. The analysis shows that public transport investment can have significant impact on the economy—around 3.7 times the amount being spent annually—and that relinquishing a car in favor of transit can save $10,103 per annum. New Zealand Transport Agency. Economic Evaluation Manual (EEM). 2016. NZTA’s Economic Evaluation Manual (EEM) describes processes for evaluating options for investing in transport infrastructure, both road and transit related. Specific sections address investment in improvements to transit infrastructure and services, including new or upgrading existing services. Investment costs must be assessed against the benefits that the changes are expected to deliver. Standard models have been developed to allow consistent analysis and comparison of different investment projects across the country. Section 4.4 (Evaluation of transport services), Appendix A1 (Discounting and present worth factors), Appendix A15 (Bus operating cost), Appendix A16 (Funding gap analysis), and Appendix A18 (Public transport user benefits) in the EEM are particularly relevant to SGR analyses.

Annotated Bibliography B-21 The approach to evaluation is based on consumer surplus methodology “to monetise the transport service user benefits of changes in price as well as non-price impacts (such as public transport journey time, reliability, frequency and comfort)” (S4.4.3). Detailed methodologies for calculating costs and benefits of investments in transit services are provided in the manual and include the following: • Benefits include transit service user benefits (such as travel time, quality of service, etc.), road traffic reduction benefits [e.g. car operating cost savings, car travel time savings (congestion reduction), accident savings], implementation disbenefits, other monetized and non-monetized impacts, and national strategic factors. • This includes provision for multimodal benefits to be considered, such as reducing the costs of road use through patronage transfer to transit. • The costs of transit services include construction, maintenance, and renewal, although these are focused on the net costs to government from whom funding is being sought for particular investments. • This is represented by the “funding gap”—where the costs of investment cannot be met via fare-box recovery, taking into account the growth of patronage that the investment is expected to generate. • Equity impacts of transit service activities should be quantified wherever possible and reported as part of the evaluation (separately from the economic efficiency calculation). Economic analysis is carried out on the costs and benefits, based on a “do- minimum” case and using defined discount rates to define time-based NPVs. Procedures are described in detail in the appendices. Bus operating costs are defined in numeric terms ($ or %) and relate to time and distance (variable costs) as well as fixed operating costs, overheads, profit margins, and capital charge (“depreciation”). The funding gap analysis process is used in the evaluation of transit service activities. It considers service provider costs and revenue, cash flow analysis, the funding gap, and sensitivity testing. The benefits described in Appendix A18 were discussed as part of the E11 project. Infrastructure and vehicle benefits, such as “Maintained,” “Ride,” and “Condition,” relate to the physical condition of the asset, which is an SGR issue, with in-vehicle time equivalents defined for each parameter.

Next: Appendix C - Model Details for Calculating Agency Costs »
Guidance for Calculating the Return on Investment in Transit State of Good Repair Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Transit state of good repair (SGR) is a critical area within the U.S. transit industry. All transit agencies, large or small, regardless of region of the country or modes operated, face challenges in maintaining their physical assets in good repair, and many are in a situation where the funds available for rehabilitating and replacing existing capital assets are insufficient for achieving SGR.

The TRB Transit Cooperative Research Program's TCRP Research Report 206: Guidance for Calculating the Return on Investment in Transit State of Good Repair addresses transit agency, user, and social costs and benefits of SGR investments. The report presents an analysis methodology that utilizes and builds upon previous research performed through the Transit Cooperative Research Program (TCRP) presented in TCRP Reports 157 and 198. The guidance (presented in Chapter 3) walks through the steps for calculating the ROI for a potential investment or set of investments.

A key product of the research is a spreadsheet tool intended for transit agency use. It is discussed in Chapter 4.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!