Tram Cb Analysis Tech Report

Technical Report Tramway Cost Benefit Analysis

Prepared for

Roosevelt Island Operation Corporation 591 Main Street Roosevelt Island, New York 10044

Prepared by

Parametrix Consulting, Inc. 1331 Seventeenth Street, Suite 606 Denver, CO 80202-1589 303-791-9235 www.parametrix.com

CITATION

Parametrix Consulting, Inc.. 2007. Technical Report Tramway Cost Benefit Analysis. Prepared by Parametrix, Denver, Colorado. December 10, 2007.

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

TABLE OF CONTENTS 1. INTRODUCTION AND SUMMARY.................................................................. 1-1 2. PURPOSE AND BACKGROUND.................................................................... 2-1 3. LIFE CYCLE COSTS....................................................................................... 3-1 4. RIDERSHIP AND REVENUE ESTIMATES ..................................................... 4-1 5. DOWN TIME ESTIMATES............................................................................... 5-1 6. NET REVENUE AND COST PER RIDER........................................................ 6-1 7. NON-QUANTIFIABLE FACTORS ................................................................... 7-1 8. CONCLUSIONS .............................................................................................. 8-1 LIST OF FIGURES 1

Annualized Cost Estimates ($ 2007) ................................................................... 3-3

2

Ridership Trends by Fiscal Year ......................................................................... 4-1

3

Fare Revenue Trends by Fiscal Year................................................................... 4-2

4

Ridership Trends by Month ................................................................................ 4-2

LIST OF TABLES 1

Annualized Cost Estimates ($ 2007) ................................................................... 3-3

2

Estimates of Annual Down Time and Replacement Times .................................. 5-1

3

Net Revenue Per Rider ....................................................................................... 6-1

4

Net Revenue Per Rider Sensitivity Analysis........................................................ 6-2

December 10, 2007 │ 354-5162-002 (01/11)

i

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

1.

INTRODUCTION AND SUMMARY This Technical Report documents the findings and conclusions of an analysis undertaken to develop information on the benefits and costs of the alternatives now under consideration by the Roosevelt Island Operating Corporation (RIOC) for the rehabilitation and/or replacement of the tramway that connects Roosevelt Island to Manhattan. This tramway has been in service for over 30 years and is now at a stage where a major investment may be required if this lifeline to the island is to remain a reliable and dependable mover of people. The approach taken in the analysis was to develop life cycle cost estimates for the alternatives under study and then compare these costs to passenger revenues and ridership estimates to gain an understanding of the relative magnitude of the cost and revenue per rider. The analysis results demonstrate that within the range of the options being considered all four alternatives would produce highly efficient investments to move people to and from the island, and that the per passenger revenue and cost differences among the alternatives are relatively small. Thus, the primary considerations in choosing an alternative become the more non-quantifiable benefits. These non-quantifiable benefits include the means of rescue in the event of system failure, system reliability and dependability, passenger service flexibility and ADA access (access for the mobility impaired).

December 10, 2007 │ 354-5162-002 (01/11)

1-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

2.

PURPOSE AND BACKGROUND Parametrix has completed detailed studies for the Roosevelt Island Operating Corporation (RIOC) to identify the best way to modernize the Roosevelt Island Tramway so that it may provide another 25 years, or more, of reliable service between Roosevelt Island and midtown Manhattan. Findings of this work are documented in a Preliminary Engineering Report prepared in March of 20071. This report identified four basic approaches to modernizing the Tramway, as follows:

Alternative 1 – Replace Critical Subsystems This option was developed for the purpose of providing a baseline, since doing nothing is not a legitimate option. This alternative includes the replacement of the track ropes, cabin hangers, cabins and track rope roller chains. Other than the required modifications, the system would remain as it is today. In the event of a major component failure, the system could be out of operation for a lengthy period of time. This alternative would make no modification to the rescue system. Therefore, in the event of a major system failure which renders the tramway inoperable, a long and arduous evacuation may be necessary. Assuming that the replacement components could be timed so that they are all available on site at once, it is estimated that the system would be out of operation for approximately 8 weeks to accomplish the rehabilitation. This alternative does not provide a high level of confidence for an additional 25 years of service. While the life expectancy of this alternative is difficult to predict, a reasonable mid-range assumption is that within 7 years additional major work would be needed to keep the tramway in reasonable operating condition. With this alternative, additional failures or issues should be expected resulting in unknown additional costs and downtimes. Alternative 2 – Replace with Similar Tramway This alternative would replace the system as it is today with approximately the same arrangement, but with the latest available technologies. The existing motors and drives would be replaced with new AC components. A new gearbox and drive train would be installed. The rescue tram would be modified or replaced to facilitate its mobilization and operation, but the general premise of a separate rescue tramway would remain. This alternative could be reasonably expected to provide another 25 years of service, or more. It is expected that the system would be inoperable for approximately 6 months during the replacement Alternative 3 – Replace with Similar Tramway, Add Redundancy This alternative is one step beyond the previous alternative. It amounts to installing a new but similar tramway, doing so with currently accepted technologies and designing additional redundancy into the system. The additional redundancy would be designed to provide a high level of availability and to potentially provide an integrated rescue approach. The additional redundancy would be achieved by installing multiple elements such two gearboxes, two AC drives and two AC motors in a way that they could be readily put into service. If the integrated rescue is achieved, the rescue tram could be eliminated. If not, the rescue system would be upgraded or replaced to improve its efficiency. This alternative could reasonably be expected to provide another 25 years, or more, of service. It is expected that the system would be out of operation for approximately 7 months during the replacement.

1

Tramway Modernization Preliminary Engineering, Roosevelt Island Operating Corporation, Parametrix Consulting, March 2007.

December 10, 2007 │ 354-5162-002 (01/11)

2-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

Alternative 4 – Replace with Dual Shuttle Operation This alternative would replace the tramway system with a dual shuttle system. Fundamentally this means having two tramways side by side, which operate independently from each other except that they share towers, terminals and operations personnel. This arrangement allows the greatest flexibility in operations and maintenance scheduling. One system may be shut down for maintenance while the other system continues to operate and serve passengers. A Dual Shuttle system could reasonably be expected to provide at least another 25 years of service and at the highest availability level of the alternatives evaluated. It is estimated that the system would be out of operation for approximately 7 months. To assist the Roosevelt Island Operating Corporation (RIOC) in reaching decisions on the most cost effective approach to the rehabilitation and/or reconstruction of the Roosevelt Island Tramway, an analysis that identifies both the costs and benefits of the alternatives currently under consideration was undertaken. Capital cost estimates for these alternatives range from $5 to $20 million ($ 2007). Of these alternatives, the long term choices are really among the last three, since Alternative 1 consists of only short-term actions necessary to address the most immediate issues. Therefore, Alternative 1 was used as the baseline for the comparison of Alternatives 2, 3 and 4. The approach to analyzing costs and benefits was performed in two steps. The first step developed an annual net revenue or cost per passenger for each alternative and was based on the best available estimates of life cycle costs, ridership and fare revenue. The approach used is consistent with similar types of analyses performed to assess alternative investments in the public transit industry. The second step then identified the potential non quantifiable benefits of each of the alternatives including the means of rescue in the event of system failure, system reliability and dependability, passenger service flexibility and ADA access (access for the mobility impaired). In all cases, Alternative 1 was used as the basis for comparison of costs and benefits.

2-2

December 10, 2007│ 354-5162-002 (01/11)

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

3.

LIFE CYCLE COSTS The development of life cycle costs was the first step in the analysis. The purpose of the life cycle cost analysis is to provide a means of identifying all costs over the life of the system and a basis for comparing the cost of alternatives that have different capital investments as well as on-going maintenance and operations costs. The end product is what is known as an annualized cost which represents the average cost per year over the expected life of the investment. For the purpose of this analysis everything was done in current year (2007) dollars. Thus, the impacts of possible future year inflation are assumed to impact costs and revenues equally. To perform the analysis estimates of capital cost, life expectancy, operating and maintenance costs, electrical power and parts and supplies were developed, as discussed in the sections that follow. The estimates, assumptions and resulting projected annualized costs are summarized in Table 1.

Capital Costs The original cost estimates for all four alternatives were developed earlier this year and are documented in the previously cited March 2007 Preliminary Engineering report. Based on the latest trends in the construction market, recent material price fluctuations, as well as the degradation of the US dollar against other currencies, the March estimates were increased by 15 percent. Given the uncertainties associated with these estimates, they are best used for comparison among the options and not for budgeting purposes. Development of an actual budget for a recommend alternative should be based on the latest available quotes in the market once more is known regarding the actual timing of component procurement and installation. Life Expectancy In order to convert the capital investment in each alternative to an annualized cost the life expectancy of the investment must be estimated. In the event that Alternative 1 is chosen, the interim repair is likely to make the tramway functional without further significant capital investment for 7 years or less. This estimate of life expectancy for Alternative 1 is a professional judgment, given the age of the system and the uncertainties surrounding the useful life of many of the existing tram’s components. Another way to look at this assumption is that the investment in Alternative 1 might keep the system running for another 7 years, at which time the system would be about where it is today in the sense of useful life expectancy going forward. The work included in Alternative 1 simply addresses the known issues with the current tramway. It makes minimal to no effort to address future, currently unknown maintenance issues. It is impossible to predict with any certainty what the next failure will be or when it might occur. Candidate components for potential failure include the gearbox, electric motor, carriage/hanger and bull wheels. Later, under the discussion of the calculation of the revenue and cost per rider, the sensitivity of the results to this life expectancy assumption for Alternative 1 is discussed. In the event that any of the system replacement alternatives are chosen, it is expected that the useful life of the system will be 30 years or greater - 30 years is a fairly standard economic life span assumption for systems of this type.

December 10, 2007 │ 354-5162-002 (01/11)

3-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

Discount Rate In addition to life expectancy an assumption regarding the time value of money or opportunity cost must be made to develop an annualized capital cost. In the public sector and given current market conditions 5 percent is a reasonable assumed discount rate and is typical of the existing cost for public sector long term financing. Operations and Maintenance (O&M) Fees The estimates of O&M fees are based on the current contract for operations and maintenance as performed by Doppelmayr. From an operations perspective, Alternatives 2 and 3 are fundamentally the same system as today, and while the new systems will likely require less corrective maintenance in the near term, no reduction was made for the maintenance fee (union labor, preference to keep “ahead” of maintenance rather than “behind,” and the redundancy creates some minor additional maintenance). Alternative 4 was increased for the anticipated staffing changes, as discussed with Armando Cordova (Operations Lead). In addition, Alternative 4 receives a credit for a reduction in the cabin attendant hours based on the possibility of using only one cabin at a time for off-peak hours Maintenance Parts and Supplies Costs The recent historical parts and supplies expenditures were used as a baseline and expanded slightly based on the recommendations of Armando Cordova (Operations Lead). To adjust this for Alternatives 2, 3 and 4, a relative estimate of near- to mid-term parts and supplies costs was made. Parts and supplies costs for Alternative 4 are higher than Alternatives 2 and 3 simply because it has more moving parts. Electrical Power Costs The recent historical power cost averages were used for Alternative 1 and as a baseline for estimating power costs for the other alternatives. Power costs for Alternatives 2 and 3 were assumed to be nominally the same as Alternative 1. Power costs for Alternative 4 were obtained by estimating the change in cabin hours under the dual system and applying that ratio to the current power usage. Additional adjustments were made to the Alternative 4 estimates to account for its different power consumption characteristics compared to the other alternatives. In summary Alternative 4 shows lower power costs compared to the other alternatives since it can operate with only a single cabin during periods of low ridership. In reality, the Alternative 4 savings may be greater because of the effect of peak power demand which should be smaller for Alternative 4. In addition, power cost estimates were increased by an additional 15 percent to reflect a rate increase that is known to RIOC Total Annual Costs Table 1 and Figure 1 illustrate the resulting estimated total annual cost for each of the alternatives. The annual cost is the sum of the annualized capital investment, annual operating and maintenance cost, annual parts and supplies costs and annual electrical power costs. As can be seen annualized costs fall into a relatively tight range from $3.4 to $3.9 million. The difference between Alternative 2, which has the lowest cost and Alternative 4, which has the highest cost, is only 15 percent. Despite having the lowest initial capital costs, Alternative 1 is not the least expensive when its short life expectancy and greater on-going costs are factored into the calculations.

3-2

December 10, 2007│ 354-5162-002 (01/11)

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

Table 1. Annualized Cost Estimates ($ 2007) Tram Alternatives 1

2

3

4

Capital Costs

$5,600,000

$14,250,000

$17,250,000

$20,400,000

Life Expectancy of Investment (years)

7

30

30

30

Down Time for Replacement

8 weeks

6 months

7 months

7 months

Discount Rate

5.00%

5.00%

5.00%

5.00%

Annualized Capital Cost

$970,000

$930,000

$1,120,000

$1,330,000

Annual Operating and Maintenance Costs

$2,300,000

$2,300,000

$2,300,000

$2,400,000

Annual Power Costs

$140,000

$140,000

$140,000

$105,000

Annual Parts and Supplies Costs

$75,000

$25,000

$25,000

$35,000

Total Annual Cost

$3,480,000

$3,390,000

$3,590,000

$3,870,000

$3.9 M

Alt 4: New Dual Shuttle

$3.6 M

Alt 3: New Similar Tram + Redundancy

$3.4 M

Alt 2: New Similar Tram

$3.5 M

Alt 1: Replace Critical Subsystems $0

$1 M

$2 M

Capital

$3 M

O&M

$4 M

Power

Parts and Supplies

Figure 1. Annualized Cost Estimates ($ 2007)

December 10, 2007 │ 354-5162-002 (01/11)

3-3

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

4.

RIDERSHIP AND REVENUE ESTIMATES Ridership and fare revenue data for the tram were analyzed for the past three years. Figures 2 and 3 illustrate the historic data and patterns for ridership and revenue by month of the year for each fiscal year. (The fiscal year for tramway operations runs from the start of the second quarter in April through the end of the first quarter in March). Figure 4 provides a more graphic look at month to month ridership changes for the last three full years of operation. For the last full year of operation from September 2006 through August 2007, the system carried 1.6 million riders and generated approximately $3.0 million in fare revenue. In addition, ridership has shown a rough 20 percent increase month to month over the past 2 years. Ridership growth reflects primarily the increased resident population on the island, which with committed development is likely to increase by another 20 percent in the next few years. As a result, the estimated baseline ridership and revenue for the analysis of costs and benefits was increased by 20 percent to a total of 1.9 million riders per year and $3.6 million in fare revenue. For purposes of the benefit and cost analysis it was assumed that the basic system capacity, ridership and fare revenue would be the same for all four alternatives. This assumption is reasonable since from a capacity standpoint Alternatives 1, 2, and 3 are the same. This assumption may understate the ultimate capacity of the dual shuttle of Alternative 4, since this system’s greater flexibility should allow slightly shorter cycle times during periods of heavy one-way passenger demand on the system. Ridership and revenue figures were then adjusted to reflect the variable down time estimates for each of the alternatives, as discussed in the following section.

180,000

140,000 120,000 100,000 80,000

2005

2006

M ar ch

y br ua r Fe

Ja nu ar y

r ov em be r D ec em be r N

O ct ob e

m be r pt e

Se

A

ug u

st

ly Ju

Ju ne

M ay

pr il

60,000 A

Riders per Month

160,000

2007

Figure 2. Ridership Trends by Fiscal Year December 10, 2007 │ 354-5162-002 (01/11)

4-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

350,000 300,000

Revenue per Month

250,000 200,000 150,000 100,000 50,000

2006

ch ar M

Fe

br

ua

ry

ry ua

m ce De

Ja n

be

be m

No

ve

ct O 2005

r

r

er ob

be em

Se pt

gu Au

r

st

ly Ju

e Ju n

ay M

Ap

ril

-

2007

Figure 3. Fare Revenue Trends by Fiscal Year

Sept. 06-Aug. 07 1.6 M Riders, $3.0 M Revenue 180,000 160,000

Riders per Month

140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 t us ug A

ly Ju

ne Ju

2005-2006

ay M

2004-2005

il pr A

ch ar M ry ua br Fe

y ar nu Ja r be em ec D r be em ov N

er ob ct O r be m te ep S

Period

2006-2007

Figure 4. Ridership Trends by Month 4-2

December 10, 2007│ 354-5162-002 (01/11)

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

5.

DOWN TIME ESTIMATES Each of the alternatives will incur periods of down time both scheduled and unscheduled. Scheduled down time for maintenance can be planned to minimize user impacts, but unscheduled down time resulting primarily from mechanical failures cannot be planned and could occur during periods of peak user demand. Table 2 summarizes the estimates for both annual down time and the length of time that the system will need to be out of service to accomplish the replacement. The annual down time estimates are based on the experience over the past few years with Alternative 1 as the baseline. Estimates for Alternatives 2, 3 and 4 reflect assessments of the degree to which down time will be reduced as a result of the newer components and the degree of system redundancy provided by each alternative. Alternative 4 shows by far the lowest estimated down time, since the dual shuttle provides two separate and independent systems, and the likelihood of system failures impacting both systems is very low. The down time estimates were used to adjust the base estimated annual ridership and revenue for each of the alternatives. Adjustments were based on a possible 7,436 annual hours of operation (20 hours per day for 5 days per week and 21.5 hours per day for 2 days per week). Given the unpredictable nature of most of the down time occurrence the average ridership per hour was used in the adjustment calculation. The replacement downtimes are as indicated in the March 2007 Tramway Modernization Preliminary Engineering report. With proper production, staging and financial incentives it should be possible to reduce the replacement downtimes stated here. The duration of and requirements for those reductions are under separate investigation. Table 2. Estimates of Annual Down Time and Replacement Times Tram Alternatives 1

2

3

4

Annual Down Time

850 hours

580 hours

440 hours

25 hours

Down Time for Replacement

8 weeks

6 months

7 months

7 months

Adjusted Annual Ridership

1.90 million

1.97 million

2.00 million

2.11 million

Adjusted Annual Revenue

$3.60 million

$3.73 million

$3.80 million

$4.00 million

Total possible annual hours of operation = 7,436 hours

December 10, 2007 │ 354-5162-002 (01/11)

5-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

6.

NET REVENUE AND COST PER RIDER

Baseline Analysis The measure of benefit cost calculated in the analysis is the net revenue or net cost per rider. Table 3 shows the results of this calculation for each of the tram alternatives. This analysis shows a small net positive revenue per rider and is reflective of the highly efficient tram operation. While the analysis implies that the tram is profitable this is somewhat misleading since there are other management and administrative costs and expenses of the Roosevelt Island Operating Corporation (RIOC) and the Metropolitan Transportation Authority (MTA) that have not been included in the calculations. These costs are a necessary component of the tram’s operation but will be the same regardless of the alternative and therefore no attempt was made to estimate them in the calculations. The results shown illustrate the relatively small differences among the alternatives with only $0.11 per rider separating the alternatives. From this work one can conclude that on a pure benefit and cost basis, as a mover of people to and from the island, all of the tram options are highly efficient investments. In addition, with relatively small differences in the results among the alternatives the decision on which investment to make will be based on other factors. Table 3. Net Revenue Per Rider Tram Alternatives 1

2

3

4

Total Annual Cost

$3,480,000

$3,390,000

$3,590,000

$3,870,000

Total Annual Revenue

$3,600,000

$3,730,000

$3,800,000

$4,000,000

Net Annual Revenue

$120,000

$340,000

$210,000

$130,000

Annual Ridership

1,900,000

1,970,000

2,000,000

2,110,000

Revenue Per Rider

$0.06

$0.17

$0.11

$0.06

Sensitivity Analysis The net revenue and cost per rider calculations are based on a number of factors as described in the earlier sections of this report. Some of these factors involve estimates or forecasts that are subject to varying levels of uncertainty and thus potentially differing outcomes. The two factors having the greatest potential variability are the life expectancy estimates and the forecasts of potential future down time for each of the alternatives. To understand the impacts on the resulting net revenue/cost calculations a sensitivity analysis was undertaken for both factors. Life Expectancy Sensitivity The base assumption is that the rehabilitation option (Alternative 1) would provide 7 years of useful life before more major work would have to be undertaken. However, it was acknowledged that given the system’s age this a professional judgment and could in reality prove too short or too long. As a result, the impact of varying the life expectancy from 5 to 10 years was tested. The other three alternatives are essentially new systems and there is no reason to expect any significant variability in their respective life expectancies. In addition, the 30 year assumption, while somewhat arbitrary, is an industry standard and of such length that varying it up or down by 5 years will have very little impact on the results of the analysis. December 10, 2007 │ 354-5162-002 (01/11)

6-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

Down Time Estimates While the hours of future system down time were forecast for each alternative based on historical information and other industry experience, the past is not a perfect predictor of the future. Therefore, it was determined desirable to understand how variation in these forecasts might impact the results of the benefit cost calculations. The baseline of 850 hours of annual down time for Alternative 1 was based on the entire data set of maintenance and service records dating back to 2001. To approximate a reasonable lower bound on the Alternative 1 downtime, events such as weather, security events and regional power outages were removed from the data set. In addition, 2004 was removed from the data set as it was recognized as not having maintenance data and 2007 was removed because it was a partial year. The long outage related to cutting the rope too short in late 2001 was also removed since this was judged to be an unlikely event in the future. After modifying the data set as described above, the historic average annual down time for Alternative 1 was found to be approximately 600 hours. To assess the impact of this change on the evaluation of Alternatives 2 and 3, the forecast down time hours for both Alternatives 2 and 3 were revised to reflect the same percentage of down time forecast for Alternative 1 in the baseline data. Specifically, the baseline Alternative 2 downtime was 580 hours, or 68 percent of Alternative 1. In the sensitivity test, Alternative 2 was forecast to experience 410 hours of down time per year, nominally 68% of Alternative 1 downtime. Likewise Alternative 3 was forecast to experience 310 hours of downtime. The down time forecast for Alternative 4 was left unchanged at 25 hours per year. Analysis Results Table 4 shows the results of the sensitivity analysis when both the life expectancy and down time estimates are varied as discussed in the previous sections. As can be seen the sensitivity analysis has the largest potential impact on the forecast for Alternative 1, with a range on the high side of $0.24 per rider revenue and on the low side of -$0.11 per rider loss, a total range of $0.35. This is reflective of the higher degree of uncertainty over both the life expectancy and reliability of the rehabilitation alternative as compared to the replacement alternatives. Alternatives 2 and 3 show a minor variation and Alternative 4 is unchanged. In total, the results of the sensitivity analysis do not change the general findings of the analysis other than to highlight the uncertainties associated with the continued operation of Alternative 1 even after the proposed rehabilitation. Table 4. Net Revenue Per Rider Sensitivity Analysis Tram Alternatives Revenue or Cost per Rider

6-2

1

2

3

4

Baseline

$0.06

$0.17

$0.11

$0.06

High Range

$0.24

$ 0.21

$0.13

$0.06

Low Range

($0.11)

$0.17

$0.11

$0.06

December 10, 2007│ 354-5162-002 (01/11)

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

7.

NON-QUANTIFIABLE FACTORS As discussed in the previous sections, it is apparent that considerations other than the cost to move people will be the influencing factors in the choice of which tramway alternative to implement. The other factors identified in the analysis include the means of rescue in the event of system failure, system reliability and dependability, passenger service flexibility and ADA access (access for the mobility impaired).

Safety / Rescue All of the alternatives meet the industry standards for passenger safety. The primary difference from a safety perspective is not in the normal operation of the system, but rather in how the failure modes are handled. In Alternatives 1 and 2, there would be no significant change in the level of redundancy in the system. Those failures today that would result in a rescue scenario would also drive a rescue scenario in Alternatives 1 and 2. Alternative 3 provides an additional level of redundancy by designing for and supplying certain “live” redundant components. For example, the design of the system could be such that in the event of a gearbox failure, within a matter of minutes, a second gearbox could be engaged and used to return the cabins to the terminals. While this redundancy does not inherently change the safety of the system under normal conditions, it dramatically changes the passenger experience in a failure mode. Alternative 4 would take this concept one step further by providing a means, under nearly any failure, of returning the cabins to the terminals. For example, it is possible with systems similar to Alternative 4 to return cabins to the terminals even in the event of haul rope sheave failure. Reliability/Dependability The tramway has become the island’s lifeline to Manhattan and a highly reliable and dependable system has great benefits to island residents. The benefit cost analysis includes consideration of the lost ridership and revenue that results when the system is down both for scheduled maintenance and unscheduled failures but does not account for the nonquantifiable benefits of having a highly reliable service. From this perspective Alternative 4 clearly performs the best since it practically eliminates both scheduled and unscheduled down time. Alternative 1 is the worst with past experience indicating that an average of 850 hours of down time will be experienced each year. While the newer components and added redundancy of Alternatives 2 and 3 show some improvement over Alternative 1 they are forecast to experience 580 and 440 hours of down time per year, respectively. Passenger Service Flexibility For purposes of this analysis, each of the Alternatives was judged to have the same passenger capacity. In general, the four Alternatives will have roughly the same effective cabin capacity and the same travel time from one station to the other. A major advantage of Alternative 4 over the other Alternatives is not a change in system capacity, nor in headways but that during times of maintenance, the system can provide 50 percent capacity. Currently, when a maintenance procedure requires that the system be out of service there is no tramway service for that period, or the maintenance cannot be performed continually. In the case of a dual shuttle system, the maintenance could be performed during low demand times, much like today, with adequate, continuous service provided by the other cabin. In addition, the dual shuttle of Alternative 4 provides greater flexibility and should allow slightly shorter cycle times during periods of heavy one-way passenger demand on the system. This in turn should allow Alternative 4 to carry slightly more people in one direction when peak passenger demands warrant. December 10, 2007 │ 354-5162-002 (01/11)

7-1

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

ADA Access Alternative 4 presents a major advantage for the Island’s mobility impaired population in that it provides for the highest level of availability. This is achieved not by a different geometry of cabins and terminals, but rather by virtue of the fact that maintenance can generally be performed on one cabin while the other cabin operates. This would greatly reduce the inconvenience to those with special accessibility needs who generally find the subway to be an inferior mode of transit.

7-2

December 10, 2007│ 354-5162-002 (01/11)

Technical Report Tramway Cost Benefit Analysis Roosevelt Island Operation Corporation

8.

CONCLUSIONS The analysis of benefits and costs for the tramway alternatives demonstrates that within the range of the options being considered all would produce highly efficient investments to move people to and from the island, and that the differences among the alternatives are relatively small. Thus, the primary considerations in choosing an alternative become the more non-quantifiable benefits. Alternative 1 provides the baseline for comparison as it is the Alternative which is closest to a “do nothing” approach. However, as described previously and in the referenced March 2007 Preliminary Engineering report, Alternative 1 does not provide the desired 25 years of reliable service between Roosevelt Island and Manhattan. In summary, using Alternative 1 as the baseline for comparison, the trade-offs become as follows:

Alternative 2 – Replace with Similar Tramway This alternative performs the best on a purely revenue per rider basis compared to all of the others. In addition compared to Alternative 1 it: ·

Reduces the likelihood of a system failure;

·

Improves reliability and dependability by reducing unscheduled down time;

·

Provides no additional flexibility improvement compared to Alternative 1 to respond to rescues should a system failure occur, work-arounds during scheduled and unscheduled outages, and peak passenger directional flow demands; and

·

Has enhanced ADA access owing to the improved reliability.

Alternative 3 – Replace with Similar Tramway, Add Redundancy This alternative reduces the per passenger revenue by 6 cents compared to Alternative 2. In addition compared to Alternative 2 it: · ·

Further reduces the likelihood of a system failure and the need for system rescue; Further improves reliability and dependability by reducing unscheduled down time;

· ·

Provides no service flexibility enhancement over Alternatives 1 and 2; and Has enhanced ADA access over both Alternatives 1 and 2 owing to improved reliability.

Alternative 4 – Replace with Dual Shuttle Operation This alternative reduces the per passenger revenue by 11 cents compared to Alternative 2. In addition compared to Alternative 2 it: · · ·

·

Virtually eliminates the need for a system rescue operation should the system fail; Further improves reliability and dependability by nearly eliminating unscheduled down time; Provides significant flexibility improvement compared to Alternatives 1, 2 and 3 to respond to rescues should a system failure occur, work-arounds during scheduled and unscheduled outages, and peak passenger directional flow demands; and Provides the most reliable and dependable ADA access of all of the alternatives.

December 10, 2007 │ 354-5162-002 (01/11)

8-1