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

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.

65 C H A P T E R 5 Introduction and Approach The guidance described in Chapter 4 of this report aids transit agencies in identifying the full impacts of state of good repair (SGR) investments versus other investment options and is useful to transit agencies of different sizes and modes. In order to test the ROI calculation guidance, a set of three pilot studies was conducted. In the summaries that follow, the pilot agencies remain anonymous but are represented geographically as follows: • Pacific Transit Agency Light Rail Line • Western Transit Agency Light Rail and Bus System • New England Transit Agency Ferry and Heavy Rail Line The results of the pilots were used to refine the guidance. This chapter describes the tools used and analyses conducted in the pilots. Analytical Approach Performing the ROI analysis involved the following basic steps: • Review the transit agency’s data, or use the Transit Asset Prioritization Tool (TAPT), to determine the scope of the SGR investments and their projected impacts. • Use the EJT Calculator described in TCRP Research Report 198 to predict EJT with and without the proposed project. • Use the Return on Investment Calculator to calculate the ROI of investing in SGR relative to deferring needed investments. TAPT and EJT Analysis Two existing tools were used in conjunction with the Return on Investment Calculator. TAPT is documented in TCRP Report 172. This tool uses information on existing transit assets and their condition to predict future SGR investment needs and prioritize investments given a budget constraint. The EJT Calculator detailed in TCRP Research Report 198 is a tool for predicting the overall journey time for a transit passenger, with adjustments for the quality of service experienced by the passenger, termed EJT. The tool also predicts changes in EJT over time based on changes in asset condition. Two versions of the tool are available: a simplified tool that does not require detailed data on journey origins and des- tinations, and a comprehensive tool that provides a more accurate prediction given origin- destination data. Pilot Summaries

66 Guidance for Calculating the Return on Investment in Transit State of Good Repair Pilot 1: Pacific Transit Agency Overview The pilot with the Pacific Transit Agency tested the ROI calculation guidance and specifically looked at investments made on a North-to-South 22-mile light rail line for refurbishments to the guideway and other assets. This light rail line, called the Pacific Rail Line in this summary, was built over 20 years ago and has assets approaching the end of their useful line. The Pacific Transit Agency initiated a project of about $91 million to restore the line’s infrastructure to SGR. The pilot tests the scenario of proceeding with the refurbishment against the base case scenario of deferring the investment. All data was acquired through public databases (such as the NTD) or from collaborating with the staff at the transit agency. The analytical approach for ROI analysis involves • Reviewing the data outlined above to determine the scope of the SGR investments and its projected impacts, • Utilizing the EJT Calculator described in TCRP Research Report 198 to predict EJT with and without the proposed project, and • Utilizing the Return on Investment Calculator to calculate the ROI of investing in SGR relative to deferring needed investments. Based on review of schedule data, the team determined that the run time for the line increased by three minutes from 2003 to 2017, in large part due to declining asset conditions. The team estimated that refurbishment of the Pacific Rail Line would reduce the effective age of the line’s guideway from 28 years to 5 years, shortening the run time to its 2003 value. EJT Analysis For this pilot the project team used the comprehensive version of the EJT Calculator to predict EJT for passengers traveling on the Pacific Rail Line. For predicting future EJT in the investment case, the run time from one end of the line to the other is reduced by three minutes and the average guideway age is reduced to five years. In the base case, the guideway continues to age but no further changes are made to running time. The fleet age is assumed to remain at 16 years in both cases. ROI Analysis Results The final step in the analysis is to enter the EJT analysis results into the Return on Investment Calculator. The results are shown in Figure 5-1. Pacific Transit Agency’s investment has a BCR of 2.33 and is projected to have a significant, positive return over the next 20 years. The fact that the rate of return is significantly greater than the discount rate and current interest rates is another indicator that the investment has a positive return. Pacific Transit Agency’s investment in the Pacific Rail Line is expected to yield benefits to the agency, to transportation users, and to society as a whole. By far the greatest benefits from the project are predicted to be agency benefits. User benefits from the investment are predicted based on the reduction in EJT. This reduction will be a benefit to passengers on the line and is predicted to result in an increase in annual trips from 22.4 to 22.6 million. User benefits resulting from these changes are predicted to be approximately $19 million in constant dollars over the 20-year analysis period. Automobile operating and travel time cost reductions are predicted to add approximately $7 million to the total user benefits. However, these benefits are partially offset by a small change in consumer surplus, which accounts for the fact that when transit passengers shift from transit to other modes besides autos they still experience some sort of travel time cost. This cost cannot be readily quantified, but the

Pilot Summaries 67 consumer surplus calculation is used so that a lost trip is not treated as a pure reduction in travel time costs. Likewise, when a passenger transfers to transit from another modes besides auto- mobile travel, the benefit they perceive from making the mode switch is approximated by the change in consumer surplus, partially offsetting the resulting increased transit travel time cost. The final category of benefits is social benefits resulting from reductions in automobile emis- sions and reductions in overall congestion from the shift of trips to transit from automobiles. Together these benefits total approximately half a million dollars. The benefits are summarized in Figure 5-2. The Return on Investment Calculator does not calculate additional economic impacts from SGR investments, but these can be estimated using the ROI calculation guidance. Refurbishment of the Pacific Line is expected to increase transit ridership by 200,000 trips per year relative to the base case in which SGR investments are deferred. The economic impact of the cost reduction from the increase in transit trips is predicted to total approximately $1.4 million annually, in constant 2012 dollars ($4.63 in personal cost savings per trip and $2.56 in business cost savings per trip). Furthermore, the spending impact of the increase of $91 million in capital investments is expected to be approximately $273 million ($3 per each dollar spent). Although it is clear from the analysis that it is worthwhile for Pacific Transit Agency to refur- bish the Pacific Line, there are additional benefits to this investment that are not fully captured through the analysis. These include, but are not limited to, factors such as the following: • The project includes upgrades and modernizations that are predicted to improve safety and reliability that may yield additional benefits not quantified here. • In the base case, Pacific Transit Agency will have a significant investment backlog as of the end of the analysis period. This backlog is included in the financial calculations, but a continuing backlog may have additional impacts on the transit agency’s ability to operate and serve its riders beyond that quantified in the analysis. • Maintaining the Pacific Line in SGR will enhance Pacific Transit Agency’s public image and level of accountability relative to allowing the line to fall into disrepair. These factors, in turn, will better enable the transit agency to make the case for needed investments in both preserv- ing and maintaining its system. Figure 5-1. ROI results for Pilot 1—Summary Measures.

68 Guidance for Calculating the Return on Investment in Transit State of Good Repair Implications for the ROI Calculation Guidance Overall, the Pacific Transit Agency pilot is extremely valuable as a test of the ROI calculation guidance and as confirmation that the guidance tends to predict a positive return for invest- ments to restore an asset to SGR. Two additional outcomes of the pilot are • The pilot served to test the use of average asset age for predicting future O&M costs. The models appear to generate reasonable results when used in this fashion. • For this pilot, an initial improvement in EJT is predicted, whereas for the other pilots either no change or a gradual change is typically predicted. The pilot served to test use of the Calculator for such a step change. Pilot 2: Western Transit Agency Overview This pilot analyzed the expected overall return of the transit agency’s investments in SGR over the 20-year period from 2018 to 2037, relative to a base case in which these investments are deferred. Specific assets included in the analysis are the light rail system (vehicles, infrastructure, and facilities), as well as transit and articulated buses (vehicles and facilities). All data was either acquired through public databases (such as the NTD) or from collaborat- ing with the Western Transit Agency. The analytical approach involved the following steps: • First, the project team reviewed the data outlined above to determine the scope of the SGR investments and its projected impacts; Figure 5-2. ROI results for Pilot 1—Benefits.

Pilot Summaries 69 • The team used TAPT to predict agency costs for the investment and base cases; • The team then used the EJT Calculator described in TCRP Research Report 198 to predict EJT for the investment and base cases; and • Finally, the team used the Return on Investment Calculator to calculate ROI of investing in SGR relative to deferring needed investments. TAPT Analysis The TAPT revenue vehicle model was used for modeling buses and light rail vehicles (LRV). This model recommends replacement based on accumulated mileage rather than age. Based on the average mileage for Western Transit Agency vehicles, bus replacement is recommended at an age of approximately 14 years, and LRV replacement is recommended at an age of approxi- mately 40 years. The TAPT age-based model was used to model rail infrastructure. Two basic types of assets are used to capture the different rail infrastructure components: track and guideway. Furthermore, track is divided into “intensive use” and default categories, with the intensive use category used to represent track in the transit agency’s central core, which is used by more traffic and thus more critical to operations. The TAPT age-based model also was used for modeling facilities. Overall, facilities were modeled rather than individual facility components. To model the investment case, the system was run without a budget constraint, which allows for making SGR investments as recommended by the TAPT models and keeping the system in SGR. A second run of the system was then performed—the base case—in which bus replace- ments are scheduled to occur four years later than recommended but all other capital invest- ments are deferred. EJT Analysis For the bus analysis, the simplified version of the EJT tool was used with defaults for all parameters, with the exception of average fleet age. An average age of 7 years (half of the useful life of 14 years predicted in TAPT) was used. The existing EJT for bus passengers is predicted to be 32.7 minutes, with 5.3 minutes buffer time, 10.9 minutes wait time, and 16.5 minutes in-vehicle time. In the investment case, this value is predicted to remain unchanged. In the base case, the average fleet age is predicted to increase from 7 to 9 years (since bus replacements will be deferred by 4 years). This is predicted to result in an increase in in-vehicle time, though other values stay the same. For the light rail analysis, the team used the comprehensive version of the EJT Calculator to predict EJT for passengers traveling on each of Western Transit Agency’s light rail lines. It was necessary to define different segments wherever lines merge or split, and then predict EJT sepa- rately for each segment. The calculation distinguishes between passengers that board at one of the stops on the segment from those that have already boarded and arrive from another segment. In the latter case, the passenger’s buffer and wait time are accounted for in the calculations for some other segment and should not be added to the total. In applying the EJT Calculator to each segment, model defaults were used for all parameters, with the following exceptions: • Specific stations for each segment and run times between stations were entered for each seg- ment using data from the published schedules; • In modeling any given segment, the team accounted for passengers boarding at a station that are part of the segment, as well as passengers arriving on a train from another segment;

70 Guidance for Calculating the Return on Investment in Transit State of Good Repair • The headway and trains per day for each segment were determined based on published schedules; and • The track type, age, and failure rate were established based on the data on the inventory in TAPT. The failure rate was set to match the number of failures predicted by TAPT, which was predicted in terms of failures per year rather than per train mile in that tool. Two future cases were modeled in the system. In the future investment case, the average fleet age was assumed to increase to 33 years, as even if vehicles are replaced on time the average age of the fleet will increase. All other parameters were held constant for this case. In the future base case, the fleet age was assumed to increase to 33 years, and the guideway age was assumed to increase by 20 years relative to the initial value given. In consultation with Western Transit Agency, the team assumed that in segments with embedded rail the run times will double. This approximates the effect a lack of repair will have on operation of the embedded rail segments. EJT model results show—for each segment—the number of daily trips originating in the segment, the number of trips transferring from another segment, the buffer time, wait time, and in-vehicle time for current conditions, the investment case, and the base case. Though buffer and in-vehicle time are not projected to increase significantly, in-vehicle time is predicted to increase as a result of the increased vehicle and guideway failures, passenger perceptions of traveling on aging vehicles, and the predicted speed reduction on the embedded rail sections. Average buffer time is 4.3 minutes per passenger and average wait time is 9.0 minutes. Average in-vehicle time is expected to increase from its current value of 19.3 minutes to 21.3 minutes in the investment case and 24.9 minutes in the base case. Travel Demand Analysis The travel demand analysis was performed by Western Transit Agency staff using information from the EJT analysis described above. They ran the travel demand model for a 20-year period for the base and investment cases using adjustment factors for bus and light rail travel time to capture the effect of changing EJT. As the time required for a transit trip increases, some portion of transit passengers may shift to other modes. Outputs obtained from the model that quantify this shift include • Daily bus and light rail trips • Automobile vehicle miles traveled (VMT) • Automobile vehicle hours traveled (VHT) Total transit ridership is projected to increase in both investment and base cases, as well as VMT and VHT. Comparing the future investment and base case, light rail trips will decrease from 46.8 million (investment) to 43.6 million (base) per year as a result of the increased travel time. Of the 3.2 million trips diverted from light rail, a portion is predicted to shift to buses. Consequently, even with the increase in bus EJT, bus trips are projected to increase from 99.0 million (invest- ment case) to 99.9 million (base case). Auto VMT is predicted to increase in the base case relative to the investment case, but auto VHT remains virtually unchanged, decreasing only slightly. ROI Analysis and Results Overall, the Western Transit Agency investments in maintaining SGR are projected to have a significant, positive return over the next 20 years. The fact that this rate is significantly greater than current interest rates is another indicator that the investment has a positive return. A summary of the ROI calculation results follows. The greatest benefits predicted from investing in maintaining SGR are agency benefits. Maintaining SGR is predicted to reduce energy costs, and reduce other O&M costs. Figures 5-3 and 5-4 show the results from the Return on Investment Calculator.

Pilot Summaries 71 Figure 5-3. ROI results for Pilot 2—Summary Measures. Figure 5-4. ROI results for Pilot 2—Benefits.

72 Guidance for Calculating the Return on Investment in Transit State of Good Repair Additional economic impacts from SGR investments are estimated using the ROI calcula- tion guidance. Investments in SGR maintenance are expected to increase transit ridership by 2.3 million trips per year, relative to the base case in which SGR investments are deferred. The economic impact of the cost reduction from the increase in transit trips is predicted to total approximately $16 million annually in constant 2012 dollars ($4.63 in personal cost savings per trip and $2.56 in business cost savings per trip). Furthermore, the expected spending impact of the increase of $766 million in capital investments is approximately $2.3 billion ($3 per each dollar spent). Many additional benefits from the Western Transit Agency’s SGR investments are not fully captured through the analysis. These include, but are not limited to, the following: • In the base case, the transit agency will have a significant investment backlog as of the end of the analysis period. This backlog is included in the financial calculations, but increasing the backlog of investment needs by over $600 million may have impacts on Western Transit Agency’s ability to operate and serve its riders beyond that quantified in the analysis. • The analysis does not consider costs and benefits geographically or demographically. For instance, deteriorating conditions of the embedded rail in the western city’s Business District may have additional economic impacts not captured in the analysis. • Maintaining the city transit system in SGR will enhance the transit agency’s public image and level of accountability. These factors, in turn, will better enable the transit agency to make the case for needed investments in both preserving and maintaining its system. Implications for ROI Calculation Guidance The Western Transit Agency pilot confirms that the ROI calculation guidance tends to predict a positive return for investments in achieving and maintaining SGR for transit assets. It tests different approaches to quantifying changes in travel demand. • The pilot, which relies on results from the agency’s regional travel demand model to quantify the impact of predicted changes in EJT on travel demand, shows it is feasible to use existing travel demand models to quantify impacts of changes in journey time as part of an SGR analysis. • In the Western Transit Agency’s case, a significant change in transit ridership is projected over time, and applying the EJT results to the model shows that there will be both a reduc- tion in light rail ridership and an increase in bus ridership in the base case relative to the investment case. The project team also performed a supplemental calculation of the benefits projected, omitting use of the city’s travel demand model. The Return on Investment Calculator then calculated changes in travel demand using the elasticity approach detailed in the ROI calculation guidance and classified the congestion impact from the change in travel demand as a social benefit. Using elasticity rather than the demand model results yields a lower estimate of user benefits: approximately $111 million rather than $132 million. The estimate of social benefits is higher: $10 million rather than $5 million. However, given the small change in benefits relative to the total, the summary financial measures are virtually unchanged. The elasticity approach also yields a different prediction of changes in ridership. In the base case results, a reduction in ridership of 2.9 million passengers is predicted relative to the investment case. By comparison, the demand model predicts a significant increase in total ridership and reduction of 2.3 million annual trips in the base case relative to the investment case. These results suggest that the elasticity estimate may provide a reasonable approximation of the results one might obtain from a travel demand model, considering just the relative change in ridership between two cases and not the overall change in ridership predicted over time.

Pilot Summaries 73 Pilot 3: New England Transit Agency Overview Two pilots were identified with a New England Transit Agency: 1) an analysis of the transit agency’s 2014 investment in Reliability-Centered Maintenance (RCM) for its Bay Line fleet and 2) an analysis of the agency’s purchase of two new ferries for its commuter boat service in 2017 and 2018. The pilot tested the scenario of proceeding with the refurbishment against the scenario of a base case, where this investment is deferred. All data was either acquired through public databases (such as the NTD) or from collabo- rating with the New England Transportation Agency. Bay Line RCM Program Pilot Scope The New England Transit Agency’s Bay Line is a heavy rail line that operates a six-mile journey between two cities. Service is provided by a fleet of 94 rail cars manufactured in 2007. Although the fleet is young relative to its expected useful life, in the early 2000s the transit agency found that fleet failures were increasing significantly. To improve asset reliability, the transit agency began an RCM program for the fleet in 2014. RCM is a data-driven approach to maintenance that is used to establish the appropriate set of time-based and condition-based maintenance treatments to perform on a complex asset to help reduce failures, maximize reliability, and minimize life-cycle costs over time. The New England Transit Agency estimates that the RCM program costs approximately $2.2 million per year, including the time spent on analysis and recommended maintenance activities. However, the program has been successful in reducing Bay Line car failures. The transit agency has observed a decline in the rate of growth of Bay Line maintenance spending, even accounting for the cost of the RCM program. Much of this reduction is due to reduced overtime charges associated with addressing in-service failures. For the Bay Line pilot, the project team analyzed the realized costs of Bay Line maintenance and compared these costs with a hypothetical alternative case in which the RCM program was not implemented. This resulted in a prediction of the realized benefits of RCM implementation. The analysis included only agency costs and benefits, as train running times and ridership have remained relatively constant over the period analyzed. Analytical Approach and Results To perform the ROI analysis, the team first analyzed Bay Line maintenance costs with and without the RCM program and then used the Return on Investment Calculator to determine the ROI for the program. Based on the analysis, the New England Transit Agency’s investment in RCM has already yielded a significant, positive return as shown in Figure 5-5. The investment is predicted to have a NPV of approximately $2.6 million and a ratio of benefits to costs of 1.29. The internal rate of return of the investment—or discount rate at which costs and benefits will be equal—is 309.75 percent. The fact that this rate is far greater than current interest rates is another indicator that the investment has a positive return. Furthermore, the payback period for the investment—or time until the sum of benefits exceeds the sum of the costs—is projected to be 2 years (reached in 2016). All of the benefits predicted are agency cost savings resulting from the reduction in rate of maintenance cost growth. The benefits are shown in Figure 5-6.

74 Guidance for Calculating the Return on Investment in Transit State of Good Repair Figure 5-5. ROI results for Pilot 3 Bay Line RCM Program—Summary Measures. Figure 5-6. ROI results for Pilot 3 Bay Line RCM Program—Benefits.

Pilot Summaries 75 The New England Transit Agency’s investment has additional benefits that are not fully captured through the analysis. The analysis is also subject to a number of assumptions. Major factors to consider include the following: • The results are driven by the rate of change observed in maintenance costs from 2013 to 2014 prior to implementation of RCM and the lower rate of growth observed following RCM implementation. Other factors besides implementation of RCM may contribute to observed trends. There is no guarantee that continued use of RCM will result in the same, lower rate of growth of maintenance costs. • There may be additional safety and reliability benefits from reducing the failure rate for Bay Line trains not captured in the analysis. • No attempt was made to calculate user benefits from RCM implementation, though these are not expected to be significant in the period analyzed. • Over time, use of best practices for maintaining the Bay Line may enhance the New England Transit Agency’s public image and level of accountability, better enabling the transit agency to make the case for needed investments in both preserving and maintaining its system. New Commuter Boat Fleet Pilot Scope The New England Transit Agency’s commuter boat service includes a set of three ferry routes. One route runs north. The other two routes run directly between two cities in the bay. The transit agency uses a fleet of ferries to provide the commuter boat service; however only two of the ferries in the fleet were owned outright by the transit agency as of 2016. These two ferries are used exclusively for the direct route between the two cities. The two together were respon- sible for approximately 47 percent of the vehicle miles reported for the system (98,341 miles out of a total of 230,426 miles). By 2016, the ferries were nearing their expected life of 20 years, and by 2018, the transit agency had purchased two new ferries. The new ferries were purchased as substitutes for the service provided by the aging ferries; however, as of 2018, the old ferries remain in service and are being used to reduce the New England Transit Agency’s reliance on the use of ferries owned or leased by others, particularly at times the transit agency ferries were being inspected or serviced. This pilot calculated the ROI of replacing the old ferries with new ones but did not address the additional benefits afforded by continuing to utilize the old ferries in a reduced role or as a reserve fleet. Analytical Approach Performing the ROI analysis involved the following basic steps: • First, the project team used TAPT to develop a ferry model and predict maintenance and fuel costs with and without the new ferries; • Next, the team used the EJT Calculator described in TCRP Research Report 198 to predict EJT with and without the new ferries; and • Finally, the team used the Return on Investment Calculator to calculate the ROI of investing in the new ferries relative to deferring replacement of the existing fleet. TAPT Analysis The project team used the TAPT revenue vehicle model to establish the useful life for the New England Transit Agency ferries and to predict fuel and maintenance costs over time. With the exception of the schedule information and cost of the new ferries, the inputs for TAPT were all derived from published NTD data. Based on the team’s inputs, the TAPT model recommended replacement of a ferry every 997,545 miles or 20 years to minimize life-cycle agency and user costs. Once the model was established, it was used to predict future maintenance

76 Guidance for Calculating the Return on Investment in Transit State of Good Repair and fuel costs for two scenarios: the investment scenario in which the existing ferries are replaced (one in 2017 and the other in 2018) and the base case in which replacement is deferred. The results for the investment case assume that the new ferries completely replace the old ferries. As discussed previously, in reality, the New England Transit Agency is keeping the old ferries in service to reduce the transit agency’s reliance on ferries owned or leased by others. The costs and benefits associated with this are not reflected in the values obtained. EJT Analysis For this pilot, the simplified version of the tool was used. The resulting EJT is predicted to be approximately 40 minutes. Replacing existing ferries reduces EJT to approximately 33 minutes, although over time EJT is projected to climb back up as the new ferries age. The differences between these values reflect the adjustment applied to time spent on an old vehicle versus a new vehicle. This increase is a function of the increased failure rate predicted if the ferries are allowed to age to 41 years, as well as further adjustments for passenger perceptions of traveling on deteriorated vehicles. ROI Analysis and Results The results of the Return on Investment Calculator are that, overall, the New England Transit Agency’s investment in new ferries is projected to have a significant, positive return over the next 20 years, as shown in Figure 5-7. The Return on Investment Calculator predicts increased ridership on the commuter boat system for the investment case relative to the base case. For the investment case, ridership is expected to grow slightly (by approximately 10,000 trips per year), while for the base case rider- ship is expected to drop (by approximately 100,000 trips per year). As there are fewer transit riders in the base case relative to the investment case, it is assumed that approximately half of the lost riders will divert to automobiles. This will reduce transit time costs but will increase vehicle travel time and operating costs. Benefits are detailed in Figure 5-8. While it is clear from the analysis that the New England Transit Agency’s purchase of two new ferries is a worthwhile investment, there are many additional benefits to investment Figure 5-7. ROI results for Pilot 3 New Commuter Boat Fleet—Summary Measures.

Pilot Summaries 77 and several caveats on the analysis. Factors to consider in interpreting the model results include the following: • Additional costs and benefits associated with keeping the existing ferries in service are not modeled in the analysis. While the existing ferries are nearing the end of their putative useful life, the New England Transit Agency could operate them efficiently for a period of at least several more years, thereby reducing reliance on use of ferries leased or owned by other entities. The transit agency may even gain modest improvements in service. • Passengers diverted from ferry service face a complicated situation not fully quantified in the modeling approach used by the Return on Investment Calculator. • The analysis does not account for the disparity in the quality of service provided by the differ- ent ferries used to serve the route. Instead, only the portion of the trips assumed to be made on transit agency-owned ferries was modeled. • Maintaining the commuter boat service in SGR will help enhance the New England Transit Agency’s public image and level of accountability. These factors, in turn, will better enable the transit agency to make the case for needed investments in both preserving and maintaining its system. Implications for the ROI Calculation Guidance The pilots demonstrate some of the additional complexities transit agencies face as they operate their transit systems and make investment decisions on a day-to-day basis. • In the Bay Line case, the New England Transit Agency faced an increase in vehicle failures for a relatively young fleet, counter to the predictions that a model such as TAPT might generate. Addressing the issues in the fleet requires in-depth analysis of the root causes of Figure 5-8. ROI results for Pilot 3 New Commuter Boat Fleet—Benefits.

78 Guidance for Calculating the Return on Investment in Transit State of Good Repair the Bay Line in-service failures, going beyond readily available summary data such as overall failure rates. • In the case of the commuter boat service, the New England Transit Agency invested in two new ferries to substitute for the service provided by two aging vessels. However, the transit agency found that rather than retiring the old ferries, it is more cost effective to use them to reduce the reliance on use of additional ferries leased or owned by others. Together, these examples suggest that the ROI calculation guidance may be extremely useful for showing the overall return on investments directed to achieve or maintain SGR versus failing to do so. However, it may not be practical to use the guidance for evaluating specific project alternatives without performing supplemental analyses to address the various complicating factors that arise in making specific project-level decisions.