Research Products and Requests for Proposals

OPTIMAL RESOURCE ALLOCATION FOR THE PURCHASE OF NEW BUSES AND THE REBUILDING OF EXISTING BUSES AS A PART OF A TRANSIT ASSET MANAGEMENT STRATEGY FOR STATE DOTS

Project 02 – 01

February 2003

Midwest Regional University Transportation Center

College of Engineering

Department of Civil and Environmental Engineering

University of Wisconsin, Madison

United States Department of Transportation

Authors:

Snehamay Khasnabis, Professor of Civil Engineering

Joseph Bartus, Graduate Research Assistant

Richard Darin Ellis, Associate Professor of Industrial and Manufacturing Engineering

Principal Investigator:

Snehamay Khasnabis, Professor of Civil Engineering

DISCLAIMER

This research was funded by the Midwest Regional University Transportation Center, and the Federal Highway Administration.  The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program, in the interest of information exchange. The U.S. Government assumes no liability for the contents or use thereof. The contents do not necessarily reflect the official views of the Midwest Regional University Transportation Center, the University of Wisconsin, or the Federal Highway Administration at the time of publication.

The United States Government assumes no liability for its contents or use thereof.  This report does not constitute a standard, specification, or regulation.

The United States Government does not endorse products or manufacturers. Trade and manufacturers’ names appear in this report only because they are considered essential to the object of the document.

EXHIBIT B

              Technical Report Documentation Page

Form DOT F 1700.7 (8-72)                         Reproduction of form and completed page is authorized.

Table of Contents                                                                              Page Number

List of Tables....................................................................................................... v

List of Figures...................................................................................................... vi

Executive Summary.............................................................................................. vii

1. Introduction...................................................................................................... 2

1.1 Background Information..................................................................... 2

1.2 Purpose of the Study.......................................................................... 3

1.3 Objectives.......................................................................................... 4

2. Literature Review............................................................................................. 6

2.1 Bus Replacement/Rehabilitation Issues................................................ 6

2.2 Asset Management Issues................................................................... 8

2.3 Use of Optimization Tools.................................................................. 11

3. Current State of Practice.................................................................................. 12

3.1 Michigan Department of Transportation (MDOT)............................... 12

3.2 Indiana State Department of Transportation (INDOT)......................... 12

3.3 Wisconsin Department of Transportation (WisDOT)........................... 13

3.4 Ohio Department of Transportation (ODOT)...................................... 14

3.5 Summary............................................................................................ 14

4. Research Methodology..................................................................................... 16

4.1 Stage 1 Process:  Computation of C_max for Rebuilding......................... 16

4.2 Stage 2 Process:  Optimization............................................................ 18

4.2(a) Model 1 Optimization Among The Four Programs.............. 19

4.2(b) Model 2 Optimal Distribution of Funds Among................... 20

                                    Transit Agencies

5. Results............................................................................................................. 24

5.1 Medium Sized-Medium Duty Buses.................................................... 24

5.1(a) Stage 1 Results (Case 1).................................................... 25

5.1(b) Stage 2 Results: Model 1 (Case 1)..................................... 26

5.1(b)i Model 1 Results: Year 2002.................................. 28

5.1(b)ii Model 1 Results: Year 2003-2009 ...................... 30

5.1(c) Stage 2 Results: Model 2 (Case 1)..................................... 35

5.1(c)i Model 2 Results: Year 2002.................................. 37

5.1(c)ii Model 2 Results: Year 2003-2009........................ 37

5.1(d) Additional Model Application (Case 2).............................. 42

5.2 Large Sized Buses.............................................................................. 47

5.2(a) Stage 1 Results.................................................................. 48

5.2(b) Stage 2 Results:  Model 1.................................................. 48

5.2(c) Stage 2 Results:  Model 2................................................... 50  

6. Analysis of Results............................................................................................ 60

6.1 Comparative Analysis of Proposed Approach..................................... 60

      and Current Practice (Case 1)

6.2 Effectiveness of the Proposed Approach............................................. 62

6.3 Comparative Analysis of Proposed Approach..................................... 66

      and Current Practice (Case 2)

6.4 Comparative Analysis of Proposed Approach..................................... 66

      and Current Practice (Case 3)

6.5 Summary............................................................................................ 66

7. Conclusions...................................................................................................... 72

8. Acknowledgments............................................................................................ 74

9. List of References............................................................................................. 76

Appendix A Supporting Tables and Figures.......................................................... A-1

Appendix B Acronyms......................................................................................... B-1

Appendix C Computation of C_max Values (For Medium Sized Bus)....................... C-1

List of Tables

Table 1             Medium Sized Bus Replacement Projections and Budget Information Obtained From MDOT

Table 2             Stage 2 – Model 1:  Allocation of Resources Among Four Program Options (REPL, REHAB1, REHAB2, REMANF) for Medium Sized Buses

Table 3             Computation of Surplus(Deficit) Resulting from 7-year Allocation for Medium Sized Buses Using C_max-values

Tables 4-7        Stage 2 – Model 2 (2002 and 2009) Distribution of RL for each agency for Medium Sized Buses Before/After Allocation of Resources (Case 1: C_max-value used)

Table 8             Large Sized Bus Replacement Projections and Budget Information Obtained From MDOT (Case 3: C_max-values)

Table 9             Stage 2 - Model 1:  Allocation of Resources Among Four Program Options (REPL, REHAB1, REHAB2, REMANF) for Large Sized Buses (Case 3 C_max-value)

Table 10           Computation of Surplus(Deficit) Resulting from 12-year Allocation for Large Sized Buses (Case 3 C_max-value)

Table 11           Analysis of Alternatives (Case 1: C_max-values)

Table 12           Comparative Analysis of Weighted Average Remaining Life for Case 1 Year 2009

Table 13           Analysis of Alternatives (Large Sized Buses C_max-value) (Case 3)

Table 14           Comparison of NPW of Improvement for Cases Computed at 5% Annual Interest

List of Figures

Figure 1            Proposed Asset Management Framework

Figure 2            Year by Year Allocation of Resources Among Four Program Options For Medium Sized Buses Using C_max-values (Case 1)

Figure 3            Year by Year Allocation Of Resources Among Four Program Options For Large Size Buses Using C_max-values (Case 3)

Figure 4            Regression Analysis for Case 1
EXECUTIVE SUMMARY  

A 1992 study conducted for the Federal Transit Administration (FTA) showed that the annual cost of replacing the nation’s transit fleet to maintain current performance levels easily exceeds $1 billion.  At a modest annual increase of 3%, this cost will exceed $1.34 billion/year.  Historically, up to 80% of this capital cost has been borne by the federal government, with the remainder being provided by state and local governments.  Current federal policies are designed to ensure that buses purchased with federal funds are kept in productive operation for a Minimum Normal Service Life (MNSL), which varies with the size of the bus.  Because of the lack of necessary capital dollars, state DOTs are generally not able to grant all requests for matching funds.  Thus, many transit agencies are faced with the task of providing service without adequate fleet size of appropriate quality.

While state DOTs may not have enough capital funds to procure new buses for its constituent agencies, it may be possible to allocate capital funds partly for the purchase of new buses, and partly for rehabilitation of existing buses, and thus meet the requests for all the buses.  Unfortunately, necessary procedures for resource allocation and asset management are not currently available in the transit literature.

In this report, the authors present an asset management strategy that is designed to enable state DOTs to (1) allocate resources for the dual purpose of purchasing new buses and rebuilding of existing buses, and (2) distribute these funds among the constituent transit agencies in an equitable manner.

The proposed approach includes two models that use the principles of optimization to accomplish the above objectives.  Model 1 attempts to maximize the weighted fleet life of the group of buses that are being purchased and rebuilt.  Model 2 attempts to maximize the Remaining Life (RL) of the entire fleet in a peer group comprising existing buses, as well as those being replaced and rebuilt.  The output from Model 1 serves as input to Model 2.

The proposed models were tested with actual planning and budgetary data provided by the Michigan Department of Transportation.  Three case studies are presented in the report that includes two cases with medium sized medium duty buses, and one with large sized buses.  For medium sized medium duty buses, there are a total of 93 agencies with a total fleet size of 720 buses to consider.  For the large sized buses, there are 21 agencies with a total fleet size of 1356 buses to consider.  For all three case studies, three rehabilitation programs are identified as alternatives to replacement (REPL) to meet the demand.  These include remanufacturing (REMANF) and two levels of rehabilitation (REHAB 1 and REHAB 2) of an eligible bus to be replaced.  Thus, a total of four program options were considered for each case study. 

In Case 1, for medium sized medium duty buses, the cost alternatives associated with each of the programs are $81,540 for REPL, $17,800 for REHAB1, $24,500 for REHAB2, and $30,320 for REMANF.  These values were used for the base year (2002) and thereafter increased by 8% every two years.  Model 1 was used for a period of 7 years, until the year 2009, to reflect the MNSL of a medium sized medium duty bus.  For each year a specific demand was met for eligible buses that needed to be replaced, within a specified budget.  The results of Model 1 indicate that a total number of 1153 buses over the period of 7 years can be “replaced” via the four programs comprising of 486 REPL, 185 REHAB 1, 63 REHAB 2 and 419 REMANF.  The total cost will be $65,054,653 exceeding the total budget of $52,889,000, resulting in a net deficit of $12,165,653 (constant dollars).  The Present Worth of this deficit (year 2002) is computed as $9,257,095, at an annual interest of 5%.  While a deficit did occur, the amount would be significantly larger had all eligible buses been replaced with new buses.

Model 2 results provide the optimal distribution of buses (funds) for the four program options among the 93 constituent agencies.  For example, an agency may have a need for 12 buses to be replaced.  However, instead of purchasing all new buses (REPL) for this agency, Model 2 may indicate that 2 should be REPL and 10 should be REHAB 1.  Model 2 considers other agency’s needs as well, and allocates funds for specific programs based on where the greatest impact in additional life will occur for the entire fleet.

For Case 2, medium sized medium duty buses were used.  However, the cost alternatives for each of the three programs were decreased.  REPL remained at $81,540, while REHAB 1, REHAB 2, and REMANF became $8,000, $12,000, and $20,000, respectively.  Using the same budgetary constraints as in Case 1, a total of 1031 buses can be “replaced” via the four programs comprising of 558 REPL, 63 REHAB 2, and 410 REMANF over the 7-year period.  The total cost amounts to $61,669,657 for a net deficit of $8,780,657 (constant dollars).  Model 2 results are different from those in Case 1.

For Case 3, large sized buses, a total fleet size of 1356 buses among 21 agencies was used.  The cost associated with each of the four programs are, REPL: $313,200, REHAB 1: $55,000, REHAB 2: $80,000, and REMANF: $100,000.  These values were also increased every two years.  A period of 12 years was used to analyze the data corresponding to the minimum normal service life of a large sized bus.  The results from Model 1 indicate that a total of 1196 buses comprising of 741 REPL, 117 REHAB 1, and 338 REMANF, can be “replaced” over the 12-year period.  With a total budget constraint of $464,660,00, the total cost of $487,421,275 results in a net deficit of $22,761,275 (constant dollars) for the period.  As in the other cases, Model 2 optimally allocates the buses (funds) among the 23 agencies so that each agency will be able to satisfy its replacement needs.                 

The case studies reiterate the fact that the budget constraints control how many buses can be replaced during any given year.  With unlimited funding, all the replacement needs for each agency can be met with new buses (REPL).  Within the budgetary constraints used in the case studies, the needs of all buses that have fulfilled their MNSL, are addressed by one of the four program options.  This results in significant improvement in the quality of the fleet, compared to the current practice of replacing only a fraction of the fleet that needs replacement.  Clearly, relaxation of the budgetary constraints (that will result in an increase in the deficit) will further improve the quality.

In all the three cases studied, the deficits incurred are compensated by the gain in the quality of the fleet, since the need for every bus that has satisfied its MNSL requirement is addressed, either through replacement or rehabilitation/remanufacturing.  The case studies also identify major shortfalls in funding and help to underscore the need for increased funding levels to improve the quality of fleet.

Results indicate that the two models are viable and require fleet life data that is typically available from state DOTs.  The application of Model 1 shows that it is possible to satisfy the fleet requirements of a group of agencies through the optimal combination of new and rebuilt buses within the constraints of a given budget.  Model 2 shows that funds for the new and rebuilt buses can be allocated among the constituent agencies in an equitable manner.  The user may specify constraints to ensure that every agency receives a minimum number of new/rebuilt buses.

1. INTRODUCTION

The addition of new buses to the existing fleet of any transit agency is a capital-intensive project. Historically, up to 80 % of this capital cost in the US has been borne by the federal government, with the remainder shared by the state and local governments. A 1992 study conducted for the Federal Transit Administration (FTA) showed that the annual cost of bus replacement needed to maintain current performance levels easily exceeds $1 billion (1).  At a modest 3% annual increase, this cost will exceed $1.43 billion today. Current federal policies are designed to ensure that buses purchased with federal funds are properly maintained and remain in productive operation for a Minimum Normal Service Life, (MNSL). The MNSL as prescribed by the FTA varies from a low of 240,000 kilometers (150,000 miles) or 5 years of service for medium light duty buses to 800,000 kilometers (500,000 miles) or 12 years of service for large, heavy-duty buses.  For medium sized medium duty buses, the corresponding figures are 320,000 kilometers (200,000 miles) or 7 years of service.

Thus, vehicles proposed to be replaced by the transit agencies must have achieved their MNSL to qualify for federal funds. If a vehicle is replaced earlier, the agency must return to FTA an amount equal to Book Value of the bus. Alternatively, an amount equal to the Book Value must be duly accounted for in the federal contribution of the newly purchased vehicle. Federal funds may also be used by transit agencies for rebuilding existing buses as well as for service expansion.

1.1 Background Information

[jb1] Most states provide matching funds to local transit agencies for the purchase of new buses through their respective Departments of Transportation (DOT); and are concerned with the escalating costs of new buses and lack of sufficient funds to keep up with replacement needs. A recent study conducted at Wayne State University (WSU) for the Michigan Department of Transportation (MDOT) investigated alternatives to replacement of existing buses (2, 3). Two generic groups of rebuilding alternatives were identified: Rehabilitation and Remanufacturing, defined as follows:

Replacement (REPL): Process of retiring an existing vehicle and procuring a completely new vehicle. Buses proposed to be replaced using federal dollars are expected to be at the end of their MNSL as described above.

Rehabilitation (REHAB): Process by which an existing bus is rebuilt to the original manufacturer’s specification. The focus of rehabilitation is on the vehicle interior and mechanical systems, including such items as rebuilding engines, transmission, brakes etc.

Remanufacturing (REMANF): Process by which the structural integrity of the bus is restored to original design standards. This includes remanufacturing the bus chassis as well as the drive train, suspension system, steering components, engine, transmission, and differential with new and/or manufactured components; and a new bus body.

The procedure developed in the aforementioned WSU study is based upon the basic principles of economics and attempts to quantify the benefits & costs associated with postponing the Replacement (REPL) of an existing bus, either by Rehabilitating or Remanufacturing, jointly referred to as Rebuilding.  The procedure incorporates tangible and intangible benefits/disbenefits, increased maintenance and repair costs for an older bus, and the risks associated with the extended use of an existing bus beyond its normal service life. The results indicate that up to certain limits, it is cost-effective to rebuild an existing bus, thereby extending its service life by another two or four years.

1.2 Purpose of the Study:

Most state DOTs receive requests for capital funds from local transit agencies for replacing existing buses that exceed the funds available.  Because of the lack of necessary capital dollars, the DOTs cannot grant all such requests under the current allocation procedures, even though the states’ responsibilities are to provide matching support of at least 20 % of the new bus cost.  Thus, many transit agencies are faced with the task of providing service without adequate fleet size of appropriate quality.

The WSU study has shown that with a fraction of the procurement cost for a new bus, it may be possible to extend the life of an existing bus, thereby addressing the problem of inadequate fleet size. Further, since rehabilitation costs are considered allowable expenses by the FTA, the fraction of the cost to be borne by the state will be small. Thus, it may be possible for the state DOT to allocate funds partly for the purchase of new buses, and partly for rehabilitation of existing buses, and still meet the requirements for all the buses.

This study devises how the state DOT should make such allocations. Clearly, this is an asset management problem that should be properly interfaced with the states' long-term strategic plans for multi-modal transportation. A procedure is not currently available that can be used to allocate a fixed number of dollars to meet the needs for given fleet size by a combination of new buses and rebuilt buses. Research reported in this study addresses the question of optimal allocation of resources for such purposes at the state level. Thus the purpose of this study is to develop a procedure that can be used by the state DOTs to optimally allocate limited capital dollars for the dual purpose of purchasing new buses and for rebuilding existing buses to be used by local transit agencies.  Additionally, a procedure is needed to allocate funds for specific programs (e.g. Replacement vs. Rebuilding), among the constituent transit agencies in the most equitable manner.

1.3 Objectives

The following are the objectives of this study:

(i)                  To identify various constraints faced by state DOTs in allocating resources for the purchase of new buses and rebuilding of existing buses

(ii)                To develop alternative procedures for allocating a fixed amount of resources for the dual purpose of purchasing new transit buses and for rebuilding of existing buses, and for allocating the resources among the constituent transit agencies.

(iii)               To validate the alternative procedures identified with case studies using actual data on fleet, needs, and cost from one of the participating states in Region 5.

(iv)              To identify the most appropriate procedure of resource allocation considering factors such as: data requirements/availability, ease of application/ interpretation, and software needs/availability.

2. LITERATURE REVIEW

Relevant literature encompasses three broad areas:  Bus Replacement/Rehabilitation issues, Asset Management issues, and Optimization Tools, as discussed below.

2.1 Bus Replacement/Rehabilitation Issues

A literature review conducted on bus REPL, REHAB and REMANF practices and policies showed that the topic had drawn significant attention in the 1980’s, and received renewed research interest in the late 1990’s.  A 1980 study conducted for the Urban Mass Transportation Administration (UMTA) by Balzer et al. examined the needs and experience of the transit industry regarding the rehabilitation or rebuilding of older buses as a means of capital improvement. (4)  Sixteen firms providing bus rehabilitation services for transit operators were reviewed.  The study recommended the use of life cycle cost analysis to evaluate alternatives for capital improvements.  This study supported bus REHAB as a supplement, rather than as a substitute, to the purchase of new buses for fleet replacement and expansion.

A 1985 study by Felicetti showed that many transit agencies use bus REHAB as a means of augmenting aging fleets (5).  The study describes various levels of REHAB, and associated costs.  The most intensive level is remanufacturing the engine and transmission, installing new bulkheads and rebuilding structural components.  The service life of a large bus for this REMANF level may be extended by ten years at a cost of $85,000.  The second level is rebuilding components and systems to original specification for added life of seven years at a cost of $70,000.  The third level includes repairs as required, with life extended up to five years at a cost of $30,000.  The report concluded that REHAB, if properly done, can be a good investment.

Bridgeman et al. in their study on the economic differences between new and rehabilitated buses noted significant variance in bus service life, operational and maintenance cost, and cost of REHAB (6).  The authors suggested use of a Monte Carlo simulation to incorporate the variance in life cycle cost analysis. This approach was not followed through however.  The authors contended that field tests that produce statistically significant, economic differentiation of new and rehabilitated buses is extremely difficult due to uncertainties associated with bus service life and operational and maintenance cost.  They suggested that significant information could be obtained by a detailed, long-term field study involving about 25 systems, each using up to 30 buses each of new and rehabilitated configuration over a period of three years.  The authors concluded that a short-term field study could provide insights into the cost-effectiveness of rehabilitated buses, even though statistical significance may not be obtained.

In another study, Bridgeman, et al., examined economic implications of new buses and rehabilitated buses, mainly focusing on different cost elements (7).  The report suggested that transit operators, when evaluating economic comparison, should also focus on other factors such as safety, passenger amenities, etc., which may have an impact on ridership.

Mcleod, et al., utilized economic analysis techniques to evaluate the effectiveness of bus REHAB from national and local viewpoints (8).  The economic factors identified by the authors to determine appropriate levels of investment between acquisition of new buses and rehabilitated buses were:  initial cost, service life, salvage, operational and maintenance cost and a discount rate.  The report describes the importance of each factor, and suggests appropriate values.  A set of formulae for estimating maximum REHAB cost and computing cost-share of FTA and local transit operators were presented.  The researchers developed a computer program to calculate maximum acceptable REHAB costs.  The report also discusses various REHAB levels, extended life of vehicles after REHAB, and economic feasibility of these REHAB levels and suggests that local operators need more funds than are currently available.  The study concluded that bus REHAB, if properly conducted, can be a cost effective alternative to new bus acquisition.

ATE conducted a study in 1985 for the Westchester County transit system in New York to evaluate rehabilitation/replacement techniques for its bus fleet (9).  A comprehensive analysis of all costs incurred over the vehicle’s life was conducted.  The operating and maintenance cost data were collected from “Operating Statistic Report” and “Expense Recovery Ratio Reports” published by APTA.  The average operating and maintenance cost figures derived from 1979 data were $1.26 and $0.40 per revenue mile respectively.  The study concluded that rehabilitation strategies such as those proposed for Westchester County, if properly conducted, can be effective in maintaining an operating fleet.  However few agencies had experience in deriving estimates for cost and benefits associated with REHAB.  For this reason, the study concluded, that REHAB is often perceived as a risky endeavor.

In another report, Bridgman et al. contends that in the mid 1980’s bus REHAB had grown in popularity due to the availability of State and Federal funds (10).  This report is a result of field study of bus REHAB programs in the United States at that time and serves as a set of guidelines rather than policies or regulations for managing bus REHAB programs.  As part of the study, the authors reviewed literature pertaining to bus rehabilitation, examined life cycle cost models and held meetings with a Bus Rehabilitation Working Group (BRWG).

According to the authors, bus REHAB is a viable alternative for fleet expansion and offers a wide range of possibilities as shown below.

The above figures represent costs for Large Bus (10 years and 350,000 miles of service), and gives an idea of the REHAB and REMANF cost.  The authors contend that the level of expenditure justified for a given vehicle must be decided on the merit of the individual case.  Further, it is impossible to provide universal guidelines on this issue.  Some operators choose REHAB of the worst bus in its fleet to extend the fleet size, while others select the best bus in its fleet for upgrading purposes and discard buses that require extensive REHAB.  The study suggested that life cycle cost analysis be used as the decision making tool to resolve the question of REHAB versus bus REPL.

A 1984 GAO Report “Bus Rehabilitation Issue Needs Attention” found that most of the funds for the purchase of new buses or REHAB of old buses were provided by the US Department of Transportation Urban Mass Transportation Administration (UMTA) (11).  The report team was aware of the growing interest of UMTA to study the costs and benefits of rehabilitating old buses and suggested the need for a comprehensive policy at the national level for bus REHAB program.  The report recommended that UMTA’s REHAB funding formula should be made identical to the formula for the bus REPL, so as not to affect transit authority’s decision to buy new buses or rehabilitate old buses.

The literature review clearly showed that remanufacturing/rehabilitating of buses, if done properly, can be a cost-effective option.  However, none of these studies considered factors such as: intangible benefits and disbenefits, increased maintenance cost of older buses and risk and uncertainties associated with the prolonged use of a bus beyond the normal life.  The aforementioned WSU study by Khasnabis, et al.. addressed some of these issues, recommended a procedure to incorporate risks/uncertainties in such decision-making, and essentially reconfirmed the studies conducted in the 1980’s (3).  No major study on the topic of allocation of capital funds for the dual purpose of REPL and REHAB/REMANF of the existing buses is reported in the literature.

2.2 Asset Management Issues

Asset management concepts have been used in the transportation field to varying degrees (12, 13, 14).  While transportation agencies manage assets, the use of optimizing techniques for transit asset management is not very common.  There is clear indication in the literature that proper use of asset management techniques can greatly enhance transportation benefits.  A trend toward using private industry principles has pushed public systems into justifying the use of resources.  Management systems have been applied to pavement, rebuilding infrastructures, human resources, bridges, traffic, and safety (15).  In the transportation field, asset management can be considered the collective grouping of all assets, including the aforementioned items.  Research efforts are currently underway to bring about the collective grouping of the component areas, as a means of asset management. The concept of resource allocation for new buses and rebuilding buses has, however, not been typically applied to an asset management program.  While the management of any specific area will change the details, asset management systems have a basic framework that can be applied to any area.

The paper Asset Management and Pavement Management (16) compares the relatively new field of asset management in transportation to the existing techniques used in pavement management.  The basis for asset management in transportation can be directly related to the mission statements of transportation agencies.  Most transportation agencies have a mission statement that includes the safe, reliable, and efficient use of transportation.  These goals can eventually be linked to the asset management system and measurement techniques as means to compile asset figures.

While no widely accepted definition for asset management has been adopted the paper presents the following:

·        “Asset management is a comprehensive business strategy employing people, information, and technology to allocate available funds effectively and efficiently among valid and competing asset needs.”

·        “Asset management is a systematic process of maintaining, upgrading, and operating physical assets cost-effectively.  In the broadest sense, the assets of a transportation agency include physical infrastructure such as pavements, bridges, and airports, as well as human resources (personnel and knowledge), equipment and materials, and other items of value such as financial capacities, right-of-way, data, computer systems, methods, technologies, and partners.”

·        “Total Asset Management is a comprehensive and structured planning process for developing capital and recurrent programs and budgets.  It aims to focus on customer and community needs, provide quality services and a commitment to excellence to ensure that assets remain productive.”

Asset Management and Pavement Management further details the commonalities between the two types of programs while also providing issues to be resolved for the further development of asset management programs.

In Asset Management Model and Systems Integration Approach (17), the authors state that the “primary goal is to achieve optimal and cost-effective investment and use of transportation asset throughout their service life cycle.”  Like the paper Asset Management and Pavement Management, this paper also states that a total comprehensive asset management system has not yet been successfully accomplished.  State DOTs use individual management systems like pavement and bridge management systems effectively. These individual components of the total asset management system used strategies to prioritize repair, replacement, etc. based on engineering and economic principles.

Asset Management Model and Systems Integration Approach presents an asset management model that is used to demonstrate key strategies and principles of asset management.  Among the key strategies are determining asset mix and inventory information and allocating asset investments.  The establishment of a baseline of information about the asset is critical for the determination of life-cycle costs.  Efficient life cycle management will eventually result in investing the right resources in the right places and the right time.  “Allocating investments between assets is at the heart of the asset management challenge,” according to the authors.  Two types of allocations exist, one with the same or similar assets, and one with different assets.  To correctly allocate resources the value of the assets must be correctly determined.  Asset management allocations are done with a combination of probabilistic, deterministic, engineering and holistic criteria (17).  However, transportation agencies are sometimes called upon to use their own judgment in the event of conflicting results by the formal techniques.

In the paper, Systems Planning for Capital Asset Management (18), New Jersey Transit’s Public Transportation Facilities and Equipment Management System (PTMS) is examined in its role as a capital programming tool.  The PTMS is one of six management systems that was originally mandated by the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA).  The act gave flexibility to states and local governments on how to best develop a PTMS for their needs.  The approach was changed in 1995 to make the PTMS a voluntary planning system.

While it is not an asset management system, the PTMS provides an inventory of existing assets with the ability to forecast future deterioration.  It aids capital programmers to make more informed decisions on how to best use the available resources.  By linking performance measures to the database, capital investment needs and strategies can be evaluated.  The modeling capabilities should also provide improved ways to compare projects (assets) and justification for those projects according to the article (18).  One of the future enhancements that the article mentions is the further development of project priorities.  A method for ranking by priority will further help capital programmers make appropriate investments.

NCHRP Web Document 41 (19) provides a comprehensive synthesis of asset management practices in the US.  Recognizing the increasing need for transportation agencies to manage assets, NCHRP Web Document 41 is a Phase I Report of a two-phase project to produce a Transportation Asset Management Guide (Phase II being the Guide itself).  This report describes recent efforts by the Federal Highway Administration (FHWA), the American Association of State Highway and Transportation Officials (AASHTO), state DOTs and others in promoting and developing asset management practices in the transportation industry.  It describes a framework for developing the Transportation Asset Management Guide and recommendations for further research in the asset management field.  Like the other documents reviewed it examines the whole asset management concept as a strategy that moves beyond the management of a single type of asset into a comprehensive approach.  Such a comprehensive approach, this report contends, may change the way the agency conducts business, reaches decisions, collects data, and communicates information.

A recent paper by Berrang et al. reports on one of the very few studies that attempt to apply the asset management concept in transit (20).  Although the term “Asset Management” is not explicitly mentioned, it attempts to explore the impact of various funding and policy scenarios on capital and operating cost consequences on the multimodal transit system operated by the Massachusetts Bay Transportation Authority (MBTA) in the Greater Boston area.  This study attempts to determine the current state of MBTA’s capital assets, and the measures needed to bring the system to a State of Ideal Repair (SIR), or “like new” operating condition.  Various strategies to attain the SIR over a 20-year period were analyzed.  The 20-year SIR was found to be sensitive to relatively small variations in annual funding levels.

2.3 Use of Optimization Tools

Mathematical programming can be used by an organization for the allocation of limited resources.  These programs usually involve either the maximization or minimization of an objective function, comprising of certain variables, which are termed as decision variables (21, 22).  The variables are then subject to various constraints, expressed in the form of inequalities or equalities.

Different optimization techniques exist, such as linear programming (LP), integer programming, non-linear programming and dynamic programming (23).  Each can be used for a specific type of problem dependent on the objective function and constraint forms.  For a LP problem, a linear relationship is assumed in the objective function between the decision variables.  Some types of problems can be formulated in the linear form except for the stipulation that the decision variables must be integers.  This type of problem is called an integer-programming problem.  Fractional answers may not make sense in some instances, and integer programming will guarantee appropriate solutions.  If the objective function or constraints are not stated as an explicit function of the decision variables, or do not fit the framework of a linear or integer combination, a non-linear approach may be necessary (22).  The term dynamic programming is a mathematic technique designed to solve complex decision making problems, where the method of solution is to categorize the problem into smaller sub-problems (22).

Our review of the transit literature shows that the use of optimization techniques in transit fleet management has been very limited, even though optimization techniques have been used successfully in computer aided scheduling and dispatching support tools for transit operation.  The framework presented in this paper uses an optimization approach for fleet management purposes.

3. CURRENT STATE OF PRACTICE

As part of this project, telephone interviews were held with a number of state DOTs in the Great Lakes Region to determine how different states allocate capital resources among its constituent transit agencies.  Brief summaries of these interviews are presented below.

3.1 Michigan Department of Transportation (MDOT)

MDOT currently does not have a formal allocation process for the purchase of new buses in place.  Federal funding for capital projects, such as the purchase of new buses, is [jb2] provided through the 5307, 5309, or 5311 programs.  If an agency receives 5307 or 5311 funds they must have met the Minimum Normal Service Life requirements for that type of vehicle.  The 20% match is provided by the state, as long as the agency receives the Federal funding.  The 5309 discretionary funds are used most frequently for the replacement of buses by agencies.  These funds are set aside for the agencies and are allocated directly to them.  There are sometimes stipulations on how these funds can be spent.  The state also receives earmarked money, which does require the state to develop some criteria to distribute the funds among various agencies.  They first eliminate any agency that received its own discretionary funds as a recipient of their funds.  They then use a criteria termed “unfunded percent of fleet” to allocate the funds among the agencies.  The unfunded percent is calculated as the number of eligible buses of a fleet available for replacement divided by the total fleet size.  A conscious effort is made by MDOT to ensure that the “unfunded percent” is below 20%. 

Each February, all the agencies must submit an annual application to MDOT for the next year’s funding.  MDOT then compiles the applications and sends them to the federal government.  As indicated earlier, MDOT generally provides the 20% non-federal match.

3.2 Indiana State Department of Transportation (INDOT)

Indiana supports approximately 48 transit agencies that are divided into four peer groups:  Large Fixed Route, Small Fixed Route, Urban Demand Response, and Rural Demand Response.  The peer groups are determined by three variables:  total vehicle miles, urbanized or non-urbanized service area, and proportion of fixed route service compared to demand response service.

INDOT, like most states, does not have a formula for capital funding allocations.  However they do use a performance-based formula to determine operating allocations.  Based on operating expenses, total vehicle miles, passenger trips, and locally derived income, this formula attempts to “create a rational and equitable mechanism for the distribution of State operation assistance to urban and rural transit providers throughout the State of Indiana” (Public Mass Transportation Funding Allocation Study).

While the direct application of the operating assistance formula may not apply to capital assistance, the methodologies and processes presented in the Public Mass Transportation Allocation Study may be applied on some level to a capital program.  These include:

·        Peer groups that reflect system size and type of service

·        Funding to each peer group based on group percentage of total operating expenses

·        Funding within each peer group based upon performance

·        Elimination of a base allocation

·        Flexibility in distribution to groups

·        Balancing the performance measures in the formula

·        Use three year rolling averages for data

·        Administrative cap for funding

3.3 Wisconsin Department of Transportation (WisDOT)

WisDOT provides state funding to support the operation of its transit systems but does not provide state funding for capital transit projects.  They use the federal funds from Section 5307, 5309, and 5311 to provide 80% of the support and the remaining 20% must come from the local agency itself.  On an annual basis the agencies will request items for capital funding.  WisDOT will verify the requests through varied means to make sure all the items requested are valid.  Wherever possible, the state administers the federal programs as a single program, usually for the Section 5309 and 5311 programs, on behalf of the agencies.  The larger agencies will apply for their Section 5307 funds directly from the federal government.

Wisconsin does have a mechanism in place for instances where the need for capital requests exceeds the amount of federal funding.  The Wisconsin State Code of Regulations, Chapter TRANS 6, describes how Rural and Small Urban Area Transportation Assistance Programs are administered.  Under Trans 6.06 rank order of priority will be given to the funds by:

            1.  Projects to replace vehicles operated by existing systems

            2.  Projects to initiate a public transportation service

3. Projects to replace maintenance and storage facilities operated by existing        systems

4.  Projects to expand the number of vehicles operated by existing systems

5. Projects to expand and rehabilitate maintenance and storage facilities of existing systems

6.  Projects to purchase and install passenger amenities for existing systems such as shelters and bus stop signs

If there are not enough available funds, the code states that “..the department (WisDOT) shall, after consultation with the applicants, reduce the scope or timing of proposed projects within that category to fit available funding.”    

Chapter Trans 8 describes the Allocation of Federal Public Transit Assistance Program Funds to Urbanized Areas Under 200,000 Population and likewise gives a similar priority ranking which shall consider age, condition, transit service needs and urgency.

3.4 Ohio Department of Transportation (ODOT)

The Office of Transit provides financial assistance to approximately 62 public transit agencies in the State of Ohio.  Of these there are 38 Rural Transit Grantees who receive capital funds each year.  The grantees are required to file a Four Year Capital and Operating Plan in December of each year for the next program year.  For example, the 2004 plan would be due in December of 2002. 

In the capital plan, the agency must list each vehicle that it is requesting to purchase and must indicate whether or not it is an expansion or replacement vehicle.  If the vehicle is a replacement, they must also indicate what vehicle it is replacing along with its inventory number, type of vehicle, age, and mileage.  This information is crucial because vehicles will only be replaced if they meet the minimum age and mileage criteria.  Exceptions are allowed if proper justification is provided (e.g. very high maintenance costs).  ODOT has been able to provide the requests for 20% funding, however in the past few years, constraints have pushed ODOT to decrease their matching capabilities.  ODOT also reserves the right to deny a request if it is not included in a previous Four Year Capital Plan without proper justification.

ODOT also has a Capital Project Evaluation and Selection Process that is utilized for Congressional statewide earmarks and the Ohio Public Transportation Grant Program.  It begins with a preliminary screening process of projects to verify they are consistent with local and statewide transportation plans.  Then the projects are evaluated, scored, and ranked for funding consideration.

3.5 Summary

The interviews conducted with different state DOTs indicated that the procedure used by the state DOTs actually vary from state to state.  However, all states that were interviewed make conscious efforts to meet the states fleet needs in different ways.  It is also clear that funding levels should be increased to meet the escalating replacement costs to maintain the quality of service.

4. RESEARCH METHODOLOGY

The methodology proposed is a two-stage process, and builds on a procedure (termed as the Stage 1 Process) developed in the aforementioned WSU Study.  Stage 1 identifies the maximum investment (C_max) that one can justify in rebuilding an existing bus when all costs and benefits are considered (2, 3).  (Note:  The term Rebuilding is used here as a generic term to denote either Rehabilitation and/or Remanufacturing as defined earlier.) 

Stage 1 is an economic analysis process that explores the economic consequences of postponing the replacement of an existing bus that has fulfilled its MNSL requirements. Stage 1 provides an estimate of C_max, the maximum amount that can be economically justified in the rebuilding of the existing bus, when all benefits and costs are considered. 

Although the Stage 1 process is already reported in the literature, a brief summary of the procedure is presented below to provide continuity.

The Stage 2 Process is a set of optimization algorithms that can be used by state DOTs to:

·        Allocate capital dollars for the dual purpose of purchasing new buses and rebuilding existing buses when the needs of all the constituent transit agencies are considered (Model 1), and

• Distribute the dollars among the constituent agencies that will enable each agency to manage their resources in an optimal manner (Model 2)

The functional characteristics of the two-stage process and their relationships are schematically shown in Figure 1.

4.1 Stage 1 Process:  Computation of C_max for Rebuilding

The Benefit Cost (B/C) method used in highway engineering and other Public Works studies expresses the comparative worth of projects by the ratio of annual benefits and annual costs.  The Benefit Cost ratio for Rebuilding (REBLD) compared to Replacement (REPL) projects can be expressed as:

(BC)_{REBLD vs. REPL} =

An interest rate i must be assumed in these computations to reflect the cost of borrowing capital. If the B/C Ratio exceeds unity, the project is considered justified. The annual benefit of REBLD is the savings in the annualized cost of purchasing a new bus as required under REPL. Thus assuming:

C_n =     Cost of new bus (REPL) ($)

n₁ =      Service Life of the new bus (years)

S_n =      Salvage Value of the bus at the end of its service life ($)

C_o =     Cost of Rebuilding (REBLD) the old bus ($)

 n₂ =     Service Life of the REBLD bus (years)

 S_o =     Salvage Value of REBLD bus at the end of its service life ($)

 K =     Additional Annual Benefits/ Disbenefits of REBLD Compared to REPL ($)

Annual Benefits of REHAB Compared to REPL

Where (A/P), (A/F) are the Capital Recovery and Sinking Fund Factors respectively at the assumed interest rate i for n₁ or n₂ years. The factor K simply incorporates all other intangible benefits (or disbenefits) associated with REHAB compared to REPL.

Annual Cost of REHAB compared to REPL

            = (Differential in Operating Cost Per Mile) * # of Annual miles * F………(3), Where:

F = Factor to reflect additional Maintenance & Repair Costs for a REHAB Bus. Thus, combining equations (2) and (3),

Annual Benefits of REHAB Compared to REPL

In equat