National Academies Press: OpenBook
« Previous: Chapter 2 - Getting Started and Negotiations
Page 33
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 33
Page 34
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 34
Page 35
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 35
Page 36
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 36
Page 37
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 37
Page 38
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 38
Page 39
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 39
Page 40
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 40
Page 41
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 41
Page 42
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 42
Page 43
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 43
Page 44
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 44
Page 45
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 45
Page 46
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 46
Page 47
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 47
Page 48
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 48
Page 49
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 49
Page 50
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 50
Page 51
Suggested Citation:"Chapter 3 - Analysis and Modeling." National Academies of Sciences, Engineering, and Medicine. 2010. Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors. Washington, DC: The National Academies Press. doi: 10.17226/14376.
×
Page 51

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.

3.1 Introduction Negotiating access to a rail corridor for a new or expanded passenger rail service often raises highly technical questions about rail line capacity, the extent and nature of capacity investments required, providing service reliability and planned journey times, estimating capital and operating costs, and determining how costs should be divided between multiple rail corridor users. In corridors with rel- atively simple operations, these questions can be answered by manual analysis (for example, using string charts) or exercising simple train performance models. However, in corridors with complex operations, either high-density traffic or operations with many long-distance and unscheduled freight trains, these questions cannot be answered by simple analyses or using a negotiator’s per- sonal experience. Instead, the negotiators need to rely on detailed analyses using the different kinds of models to resolve these questions. Modeling methods and model inputs must be acceptable to all parties to ensure that the results are trusted and can be used to guide the parties toward an agree- ment. It is generally not productive for each party in the negotiation to perform its own analysis and compare results. Experience has shown that the results from different models are rarely identical. They can differ enough to cause disputes about the validity of the models for the specific situation being analyzed—the unproductive “dueling models” situation. This situation can add to negotiat- ing difficulties rather than guide the parties toward a mutually acceptable agreement. The specific areas where detailed analysis and modeling may be used include: • Operations simulation and capacity analysis. • Capital cost estimating and cost sharing. • Operating costs and cost sharing – Amtrak avoidable-cost methodology. – Fully allocated costs for commuter rail. The following sections discuss analysis and modeling techniques used for passenger projects and how they have been applied. 3.2 Operations Simulation and Capacity Modeling 3.2.1 Simulation and Modeling Overview Except for the simplest of operations, there is no easy formula that will yield the capacity of a rail corridor. There are simply too many variables involved (train characteristics, speed limits, train and siding lengths, signal system characteristics, etc.) for any simple approach to yield use- ful results. Because of this level of complexity, all methods of operations simulation and capac- ity analysis rely on detailed operations simulation methods. 33 C H A P T E R 3 Analysis and Modeling

Simple Models In its simplest form, a simulation model is a computer program that performs a stepwise cal- culation of the movement of a train over a rail corridor. Using information on speed limits, grades, train acceleration and braking rates, station stop dwell times, etc., the model calculates the speed and distance traveled by the train for each time step (e.g., every 10 seconds). After the model has stepped along the whole corridor, it produces a tabulation of time and distance traveled, often presented graphically as a time vs. distance string-line chart. A model that performs this calcu- lation for a single train moving over a rail corridor is usually known as a Train Performance Calculator (TPC), because it calculates travel time without interference from other trains oper- ating on the corridor at the same time. TPCs often have additional features, such as an ability to calculate energy used or fuel consumption. Single-train TPC calculations are used to determine what rail corridor upgrades will be required and to provide the desired travel time before the interference effects from other trains and other typical operating delays are taken into account. For initial planning, it is customary to pad the minimum trip time by around 10 percent to esti- mate a practical trip time. This type of calculation can be used to investigate such questions as the reduction in journey time from increasing top speed from 79 mph to 110 mph, or adding or omitting station stops. Complex Models The more complex version of a train operations simulator performs a simultaneous calcula- tion of all train movements on the corridor, taking into account signal system characteristics, train priorities, temporary slow orders, and typical dispatcher decisions over where trains should meet or overtake each other. At their most complex, the multi-train simulations closely reproduce how a real rail corridor would be operated, taking into account all the variations in individual train performance and other operating constraints and variations. Results are usually presented as the calculated trip time for each train compared with minimum time with no interference from other trains, slow orders, etc. The difference is reported as a delay. Operation over the corridor can also be represented on a string-line chart (see Figure 3-1) or as an on-screen animation— a speeded-up version of a dispatcher’s display. However, a single run of a corridor operations simulation will only represent operations under one set of input conditions. Railroad operations are subject to a variety of random and planned disruptions to normal operation, including planned and unplanned track maintenance, delays at stations, and delays caused by events elsewhere on the railroad. Freight train operations are not normally conducted with great precision, and even scheduled freight trains are subject to variability. In addition, many through freight trains are unscheduled “extras” that run as needed and may enter the corridor at any time. Multiple model runs are used to address these variables, with results presented as average run times and delay statistics for each train, along with string charts and animations as required. The primary use of a multi-train simulation model is to investigate which infrastructure upgrades to an existing rail corridor are needed to enable it to accommodate additional passen- ger train trips while still meeting specified service performance requirements (train departure times, trip time, and on-time performance) and complying with any other specified constraints. The analyst will start with improvements identified using a single-train TPC (if available) and will make multiple model runs to test alternative track configurations and other improvements. The objective is to identify a cost-effective package of improvements that will meet the service requirements of all users. Given the trial-and-error process of using simulation models, the com- plexity of these models, and the potentially large number of alternative corridor configurations to be investigated, an experienced modeling analyst is essential. Modeling is something of an art, and a model cannot represent everything about a route. Interpreting results requires judgment, 34 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

Tuesday (Day 2) Sunday (Day 0) Monday(Day 1) Wednesday (Day 3) Note: Trains 1-6 are passenger and the remainder are freight. Source: UVA IDOT Base Case report for the FRA Figure 3-1. String charts for different days of the week on a segment of the Chicago to St. Louis corridor.

informed by experience using the model, experience with interpreting the results, and experi- ence observing real-life outcomes. 3.2.2 Choice and Availability of Models This discussion focuses on appropriate models to support detailed negotiations with a host railroad to operate passenger train service over a busy rail corridor. The model used must be able to take into account all operating constraints and corridor features and must be accepted as pro- ducing valid results by all parties in the negotiation. Simpler or alternative models are often used during earlier planning stages and for corridors with less complex operations. In many cases, these requirements limit the choice to two comprehensive rail operations sim- ulation software packages that are widely used in the railroad industry for capacity evaluation and project and service planning. These packages have been used in almost all recent passenger rail service developments where corridor operations were sufficiently complex to warrant the use of detailed modeling. Some agencies may be uncomfortable with the apparently restricted choice of simulation soft- ware for capacity analysis and will point out that other models exist and have been used success- fully in the past. Although this is true, the firms responsible for the two leading software packages have continually refined their products, added new features, and built a broad user base among passenger and freight railroads. These efforts have given the models an industry-leading posi- tion, overtaking the competitors. In addition and most important, many passenger and freight railroads have invested considerable resources in assembling rail network infrastructure and operations data for use with their chosen package. It is costly and time consuming to re-input these data for use with another package, even if one were available and would be acceptable to all parties. The advantage of using one of the leading simulation models is that the results are highly realistic and are likely to be accepted as such by all parties. The long history of successful applications of the models to different corridors means that most potential “bugs” have been found and corrected, so the parties are very unlikely to encounter surprises when the new ser- vice starts operation. For these reasons, usually the most practical option regarding model choice will be to use the train operations software package routinely used by the host railroad to ana- lyze operations on the corridor. The two packages are Rail Traffic Controller (RTC) developed by Berkeley Simulation Software and RAILSIM developed by the rail consulting firm SYSTRA. RTC is specifically designed for application to North American freight railroads, with substantial unscheduled train movements and a range of signaling and train control methods. RAILSIM is most commonly applied to higher-density passenger rail and rail transit corridors and contains modules for modeling pas- senger car fleet utilization and other service design aids. In spite of this specialization, either pack- age can be applied successfully to passenger-dominant or freight-dominant shared corridors. The rail operations simulation models used by both packages are data-intensive and relatively costly to use. They require detailed information on track layout, speed limits, signal systems (including signal block lengths), and full details regarding the rail traffic operated. This detail includes data for each train or type of train operated, including locomotive or traction power and braking characteristics. The large amount of data needed, however, make these packages cumber- some and not well suited for use in earlier-stage planning studies or for the rapid screening of mul- tiple alternatives. Simpler models have been developed for this purpose, such as the Association of American Railroads Train Energy Model (TEM), a train performance calculator particularly designed to evaluate the effects of route characteristics and operating techniques on trip time and energy consumption. Also, several commercial consulting firms, such as Parsons Brinckerhoff and Zeta-Tech, have developed screening-level models for capacity and train performance analysis. 36 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

Rail Traffic Controller Software Package RTC simulates train movements over a rail corridor by mimicking the behavior of a skilled dispatcher, given rail traffic volume and mix, track layout, signal system, etc. A TPC module cal- culates train progress over the corridor in response to signal indications and dispatcher commands. Results of each model run are presented as traffic flow animations, time-distance (string-line) charts, train delay statistics (the primary measure of whether capacity is adequate), TPC outputs, and various train operating statistics, such as average speed (velocity). The model can be used to evaluate the sensitivity of corridor performance to traffic level and mix, evaluate benefits from capital improvements, design a train schedule, and identify bottlenecks or capacity constraints and options for removing them. Almost all the U.S. Class 1 railroads have standardized on RTC and have prepared infrastructure and operations data for entry into RTC for much of their route network. The principal advantage of the RTC package for freight railroads is that it is specifically designed for North American freight operations and fully accounts for the characteristics of such operations. It has been adapted to user experience to make it the leader for U.S.-style freight operations. RAILSIM Software Package RAILSIM is the second widely used software package for train operations simulation. Applications of this software have mainly been to higher-density and complex commuter and rail transit operations, to plan optimum track and signal system layouts, and to minimize delay potential and energy consumption. RAILSIM was recently used to plan Caltrain commuter ser- vice developments between San Francisco and San Jose, California, a predominantly passenger corridor with limited local freight service. It is also widely used in the New York City region on predominately commuter corridors, such as the Long Island Railroad, Metro-North Commuter Railroad, and New Jersey Transit, and in the Chicago area by Chicago Metropolitan Rail (METRA) and Northern Indiana Commuter Transportation District (NICTD). 3.2.3 Using Simulation and Modeling in Rail Corridor Planning and Negotiations The recommended approach to applying rail operations simulation, analyzing the capacity of a shared corridor, and determining the infrastructure investments needed to provide a defined freight and passenger service is discussed in the following subsections. Step 1. Agree on What Type of Analysis is Required and What Model to Use The first step in any operations simulation modeling is to agree on what kind of analysis is required and what simulation package will be used. The kind of analysis will depend on the com- plexity of operations on the corridor. A corridor with simple operations—such as four or five daily passenger round trips, a daily local freight train, and three or four through freight trains that run to a predictable schedule—should not need complex modeling. Instead, a simple train performance calculation and string-line time-distance plots should be sufficient to plan passing siding locations and where track upgrades to increase speeds are worthwhile. A corridor with more trains, trains with different priorities, and long-distance through trains with unpredictable schedules will require more advanced analysis. As indicated earlier, it will be most cost and time effective to use the train operations software package routinely used by the host railroad to analyze operations on the corridor. The freight railroad will likely already have prepared train and infrastructure data for the model, greatly reducing the effort needed to prepare for analysis. Almost all Class 1 freight railroads have stan- dardized on the RTC model, so it is highly likely that the best choice for capacity and infrastruc- ture investment analysis for a new passenger rail service on a busy Class 1 rail corridor will be the Analysis and Modeling 37

RTC package. Stakeholders report that freight railroads consider RTC to give the most reliable capacity analysis available and are very reluctant to accept results from alternative models as a basis for estimating capacity requirements of a proposed passenger service. Passenger rail agen- cies that have participated in RTC-based capacity analysis report that they have been satisfied with the results and the resulting decisions regarding the need for capacity investments. The situation will be similar on a busy commuter rail corridor (often used by an intercity pas- senger service for final miles into a city center), except that the software package of choice is more likely to be RAILSIM. Smaller host railroads (for example, a non-Class 1 freight rail or a smaller commuter operation) will likely not have previously conducted detailed operations simulation analyses; if detailed analysis is necessary, the parties are more free to choose among the available analysis packages. In all cases, this Guidebook strongly recommends that a single series of capacity and infra- structure analyses are performed using the selected analysis package. Both host and tenant should collaborate on specifying analysis inputs and in reviewing and interpreting results. Step 2. Identify Passenger Rail Agency Analysis Team and Agree on Modeling Procedures The passenger rail agency must enter discussions with a host railroad on capacity and infrastruc- ture investments with adequate technical support. Most important, the agency’s team must include an individual with experience with the specific operations simulation software selected for capacity analysis. In many cases, the analysis will be carried out using the host railroad’s software and corri- dor infrastructure input data. To ensure the passenger rail agency’s interests are fully represented in the negotiations, model inputs and results must be subject to a knowledgeable independent review. When implementing passenger service on a major freight railroad, the most widely recom- mended approach to performing evaluations of a proposed passenger service is for the railroad and passenger rail agency to agree that one party (the freight railroad, the passenger rail agency, or a mutually acceptable consultant) will perform the analysis and share inputs and results with both parties. According to reports, some railroads have insisted on doing the calculations themselves and have been reluctant to share some details of inputs and results. This reluctance may be because of concern about revealing confidential business information or caution about getting involved in a lengthy dispute about the validity of input assumptions or interpretation of results. However, as confidence has grown in cooperative analysis and with the use of confidentiality agreements, these concerns have faded. Stakeholders report that the modeling process is becoming much more open. Step 3. Agree on Key Rail Service Inputs Once the analysis package has been selected, the next step is to define the passenger and freight rail services that will operate over the corridor. The typical data needed are discussed in the following paragraphs. Passenger Service. The service data needed for either intercity or commuter service or both will include: • Target trip time, with station stops as planned. Both trip time without delays and a practical schedule time with an allowance or padding for delays should be defined. Adding excessive padding (such as exceeding 5 minutes per hour) to cover for poor operations or maintenance procedures should be avoided. • Station locations and expected station stop dwell times. • Train departure times and the number of daily trips in each direction. • Service quality parameters, primarily OTP at final and intermediate stations, and fraction of trips missed due to equipment or infrastructure problems. 38 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

If the service is to be developed in stages, this information is required for each develop- ment stage. Freight Service. The service data needed for freight service includes the numbers and types of trains forecast to move over the corridor corresponding to each passenger service develop- ment stage. Numbers of trains will be based on the railroads’ business plans and economic fore- casts. Arrival times of trains approaching the corridor should be based on the freight railroad’s operations plans, with a measure of the variability of arrival times. Some trains will follow pre- dictable schedules, for example, intermodal trains and regular local freight trains, while some may be seasonal, such as grain trains that run during harvest seasons. Some trains will run com- pletely at random. Day-of-week variations are also typical in freight service. It is advisable for the passenger rail agency to review the freight traffic forecasts independently, and raise questions if these appear to be inconsistent with state rail plans, likely local and national economic condi- tions, and announced plans to change freight train routing. Step 4. Carry out the Analyses The analyses will be tailored to each specific passenger rail situation. However, a typical set of analysis sequences might include: • TPC-only analysis for passenger train trips to establish minimum track upgrades and speed increases to achieve the desired trip time, after allowing for reasonable schedule pad. • Analyses of present-day freight and passenger (if any) operations on the corridor to check that the analysis adequately represents the real world and also to establish a baseline for the cur- rent performance of trains using the corridor. In particular, current delay statistics on the cor- ridor will set a target for corridor operations with added passenger trains. • Further sets of simulations analyses with corridor speed upgrades determined using the TPC, and various capacity investments (double track, passing sidings, train control systems) to identify improvements that meet both capacity and performance goals. Delay statistics are the primary measure of performance; the corridor must maintain at least present freight train per- formance levels while meeting passenger train targets. This analysis is repeated for each stage in passenger service development if a multi-stage program is planned. • Joint review of each analysis by the passenger rail agency and the host railroad, and discussion and resolution of any questions. Case Studies 4 provides an example of ongoing use of simulation modeling (in this case using RTC models) by the Capitol Corridor Joint Powers Authority (CCJPA) in California to manage track and service improvements on this corridor. CASE STUDIES 4 Application of the Rail Traffic Control Model and Capital Planning The CCJPA and its host railroad, the Union Pacific (UP), make extensive use of the RTC model for ongoing planning of capital investments. Full details of the route and passenger and freight services are coded into the model. CCJPA develops a long-term “Vision Plan” for the corridor and communicates this to UP. UP then runs the RTC model and develops plans for investments needed to accommodate each increment of passenger rail service, especially identifying bottlenecks that would constrain future freight service growth and journey time reductions. UP also develops cost estimates for the proposed improvements so that, after review and negotiations, CCJPA can issue a work order under the master agreement to execute the work as funding becomes available. Analysis and Modeling 39

3.3 Capital Investment Planning, Costing, and Cost Sharing Almost all proposals for new or expanded passenger service require capital investment in infrastructure along the route, either to improve the quality of the infrastructure, to support the proposed passenger service, or to add needed capacity. The steps in this investment process are first to determine what is needed to support the proposed passenger service, then to estimate the capital cost of the required projects, and finally to determine how these costs will be shared between the freight railroad and the passenger rail agency. The details of a capital project also have significant implications for ongoing maintenance costs. Adding a passing siding means that maintenance cost will increase. Rebuilding deteriorated track structure with new rails, ties, and ballast will reduce ongoing maintenance costs for a period of time. 3.3.1 Right-of-Way Access or Acquisition The different approaches to gaining access to a rail corridor for a new or expanded passenger rail service are discussed in Section 2.4.2. This section discusses methods to place a price on gain- ing access and to estimate the cost of completing corridor upgrades to support the planned pas- senger rail service. The factors to consider in estimating costs and prices are discussed in the following subsections. Amtrak Intercity Service Amtrak’s right of access means that it can make use of existing available capacity on a rail cor- ridor at no cost and is only obligated to pay incremental costs for use of a host railroad’s tracks. However, in many cases, the existing infrastructure is not adequate for a planned service, either because track condition and signal system capabilities will not support planned train speeds and/or because there is insufficient capacity to accommodate planned passenger and freight oper- ations. The need for additional capacity should be based on objective analyses as described in Section 3.2, to ensure that the corridor is able to accommodate planned and forecast passenger and freight services with planned investment. If necessary, a passenger rail agency will need to negotiate funding for track and signal system improvements and additional line capacity needed for passenger service. The host railroad would be responsible for any investment that it would have needed to accommodate forecast traffic, if the passenger service had not been implemented. The process for estimating capital costs of track and signal system improvements and additions, and shares of cost to be borne by host and tenant are discussed in separate sections. In a few instances, it may be attractive for the passenger rail agency to purchase a rail corridor or part of a rail corridor, usually where freight traffic is very low and there is a willing seller. This approach has not been used in the past—opportunities were few and funds were lacking. However, outside the NEC, Amtrak does operate over portions of several commuter networks, owns a segment of the Chicago to Detroit corridor, and operates a corridor in North Carolina owned by the state. Commuter Service Commuter rail agencies do not have Amtrak’s right of access to the rail network and there- fore must expect to incur a cost for access to a rail corridor that is separate from capital expenses for infrastructure improvement and a share of operations and maintenance costs. The different approaches used by passenger rail agencies and their cost implications include: Purchase the Rail Corridor. This option is commonly selected for commuter rail, where the corridor was a low-traffic freight branch line prior to introducing a new service. The value of a corridor as a freight rail business or the value of railroad materials installed in the corridor is gen- 40 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

erally quite low, and the sale price is primarily a function of the corridor’s real estate value. This value depends on many local factors and would be beyond the scope of this Guidebook to sug- gest methodologies to estimate a fair purchase price. In most cases, the track and signal systems (if any) have to be completely replaced. The rail freight operator usually is given permanent access to the corridor to provide ongoing service, subject to time-of-day restrictions so that freight operations do not interfere with the proposed passenger operation. Directly Fund Infrastructure Improvements. Commuter rail agencies can directly fund infrastructure improvements sufficient to provide the capacity consumed by the passenger rail service, plus any signaling and track quality improvements required for passenger service. This approach would be used where the host railroad is willing to accommodate the passenger ser- vice but requires that existing capacity for freight operations is preserved, including maintaining spare capacity for future freight growth. A freight railroad may also require that capacity is main- tained at specific times of day, depending on the type of freight traffic using the corridor. In other instances, the commuter and freight railroads have agreed on a combined package of invest- ments, with costs shared, to serve passenger service requirement as well as anticipated freight traffic growth. Every case is different and must be worked out in negotiations. There are several examples of this approach, such as improvements on Norfolk Southern Railway and CSX in Northern Virginia for the Virginia Railway Express (VRE) commuter services into Washington, D.C. Often the improvements are not a one-time investment but are a series of projects imple- mented as required by passenger and freight traffic growth. Purchase a Permanent Easement. A permanent easement can be purchased for a speci- fied commuter rail service (number of trips, schedule, journey time, etc.). This option has two sub-options: • Acquire an easement to construct a separate commuter rail line parallel to a freight line within the existing railroad ROW. This option is selected when it is not technically feasible or is very difficult for the commuter rail service to share track with the freight operation, and lateral space exists within the existing ROW. The commuter rail agency is responsible for construct- ing the commuter tracks, signal systems, stations, etc. This option has been used for the Gold commuter rail line in Denver and the Front Runner service north from Salt Lake City. As with outright purchase of a rail corridor, price will depend on local circumstances. • Purchase a permanent easement to operate commuter rail service on the tracks of the host rail- road for a lump sum. The easement seller guarantees access for a specified number of passenger rail trips to specified schedules and trip times and is responsible for making infrastructure invest- ments to honor this agreement. This option bundles the cost of access with the cost of infrastruc- ture improvement. Because of the bundling, and the permanence of the agreement, it may be difficult to compare the lump sum payment with alternative estimates of access and infrastruc- ture improvement costs, and to justify the payment to funding agencies. The primary example of this approach is the Sounder commuter service northward from Seattle. As well as these two sub-options, access and infrastructure improvement costs can be con- verted into a per-train-mile usage cost to be paid as incurred. This approach was used for the Sounder service between Seattle and Tacoma in Washington State. Hybrid arrangements are also possible, using some mix of public funding of infrastructure improvements, a one-time capital payment, and ongoing per-mile charges. 3.3.2 Estimating Capital Costs The majority of corridor access arrangements, whether for Amtrak intercity or commuter service, place the responsibility for funding and implementing corridor infrastructure improve- ments on the passenger rail agency developing the service. It is assumed that the passenger rail Analysis and Modeling 41

agency and the host railroad will have performed capacity analyses as described in Section 3.2 and have determined what improvements are required to accommodate the planned passenger service as well as freight traffic growth. The improvement plans will reflect: • Service quality and performance requirements for both passenger and freight operations. • Improvements in track quality to meet requirements for planned passenger train speeds. • Signal and train control system investments to comply with regulatory requirements, espe- cially a PTC system suitable for planned speeds and able to provide the required line capacity. Some forms of PTC may not meet regulatory requirements for operations over 79 mph and also may reduce rather than increase capacity. • Investments to increase capacity, such as adding or lengthening passing sidings, adding a sec- ond track, or changing signal block spacing. • Any more detailed factors that affect capacity, such as a need to provide capacity for specified freight operations during passenger service peak period or to maintain total daily freight capacity while restricting freight train movements through commuter stations during peak travel hours for safety reasons. • Investments in passenger station infrastructure, including purchasing or leasing land for the station and constructing the station. In many cases, a station will host multiple public trans- portation services to serve as a transportation hub. In such cases, station design and cost shar- ing will have to be worked out with other agencies sharing the facility. Once physical requirements have been defined, the next step is to estimate the cost of imple- menting the improvements and to plan the construction work. Activities and responsibilities of this step will depend on whether the construction is on a corridor that has been purchased or leased by the passenger rail agency or is on an active host freight railroad. On a corridor owned by the passenger rail agency, the agency can retain an engineering firm to prepare detail designs for the improvements, estimate costs, and select an experienced contractor to perform the construction work. The contractor and the agency will have to work with any ten- ant freight railroad to schedule construction work so that rail freight service can continue. Cost estimates must take into account the impact of accommodating ongoing freight and passenger operations, where applicable. On a corridor owned by an active freight railroad, the host freight railroad will develop a cost estimate (taking into account the need to work around ongoing freight operations) and manage construction under a contract with the passenger rail agency. A freight railroad may retain an experienced engineering firm to do the cost estimates and design work, depending on the work- load in the railroad’s engineering departments and the nature of the work. Similarly, the railroad will determine whether to use contractors or its engineering department employees and equip- ment for the construction or upgrading work. Railroad union agreements often govern the con- ditions under which contractors may be used. As the source of funding for these projects, the passenger rail agency must review the railroad’s cost estimates for reasonableness and properly inspect completed work before releasing payments. 3.3.3 Sharing Infrastructure Capital Costs Many passenger rail agencies understand that a host railroad will obtain some benefits from infrastructure improvements to accommodate a proposed passenger rail service. Because there are identifiable benefits for the host railroad from these investments, passenger rail agencies believe that the freight railroad should contribute funds toward to the cost. There are two situ- ations where cost sharing will be on the table in negotiations: • Where investment in the corridor is already under consideration in the railroads planning. For example, a railroad may plan to replace an older automatic block signal (ABS) system with 42 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

Centralized Traffic Control (CTC) and replace manual siding switches with powered switches controlled from the central office, but the project is not sufficiently beneficial or urgent for the railroad to implement on its own. A contribution from the passenger rail agency will reduce the total cost to the railroad, enabling the project to go ahead under a cost-sharing agreement. • Where the proposed passenger corridor is not an investment priority for the freight railroad, but the freight railroad will receive identifiable benefits from passenger-related investments. Stakeholders in passenger rail agencies often express the view that the freight railroad should share in the cost of these improvements, because it will receive benefits in the form of higher speeds, fewer delays, quicker trip times, etc. The benefits can be estimated using the results from capacity modeling. However, it can be difficult to reach an agreement on whether these benefits have any value to the freight railroad and, if so, how to determine the value. The difficulty derives from how a freight railroad, or indeed any private business, approaches capital planning. At the beginning of its fiscal year, a railroad will estimate the capital expendi- ture it can afford in the year based on estimated earnings, borrowing costs, and related factors. Some investments will be to support long-term strategies established by the railroads’ top man- agement; some will support previous commitments, such as a contract to purchase a number of new locomotives that year; and others will be selected from among candidate projects submit- ted by the railroad’s operating, engineering, and mechanical departments. In the latter case, proj- ects will be ranked by return on investment (ROI), and the railroad will fund those with the highest ROI, up to the total of available funds. Projects that do not make the ROI cut will not be funded. It is unlikely that a railroad will be willing to contribute to a passenger service–driven investment if that project would not otherwise qualify for an investment, given the railroad’s pri- orities and the current ROI hurdle rate. Alternatively, a freight railroad might be willing to con- tribute, provided the corridor is an important link in its network and the benefits relative to the investment meet the railroad’s ROI and other investment criteria. The best approach to reaching agreement on capital cost sharing is to discuss at the beginning of the capacity analysis what the railroads investment service needs on the corridor might be and seek to structure the project to meet both freight and passenger service needs. If the corridor is part of the railroad’s core network, then the freight railroad would likely be willing to contribute at a level where the benefits yield an adequate ROI. If the corridor is not a priority for the freight railroad, then the passenger agency will likely have to fully fund the improvement project. Another cost-sharing possibility occurs where a state agency is able to contribute funds from freight rail and grade crossing safety improvement programs. This was the case in Washington State on portions of the Portland, Oregon, to Seattle, Washington, and Sounder commuter ser- vice lines. Even if the state is able to provide funds from a freight rail program, the investment in the corridor must still be of value to the rail freight business and more broadly contribute to state freight transportation and economic development objectives. 3.3.4 Rolling Stock Capital Costs Any new service has to provide for equipment (passenger cars and locomotives) to operate the service. There is no one best or recommended approach—use whatever meets objectives for the specific corridor and service at the time the equipment is needed. A summary of the approaches to providing equipment and the implications for capital costs follows: • Amtrak Intercity Service – Purchase new equipment. In this case, a competent railroad rolling stock consultant can develop specifications for the trains and develop a cost estimate based on recent prices paid to suppliers of similar equipment. Analysis and Modeling 43

– Lease new equipment. This approach converts the capital costs into a periodic payment, which becomes part of ongoing expenses. Whether leasing is attractive to a public agency depends on how accounting practices and funding arrangements treat capital leases. Discussion of the financial implications of leasing is beyond the scope of this Guidebook. In some cases, a capital lease may be bundled with equipment maintenance services. – Fund rehabilitation of existing out-of-service cars in Amtrak’s fleet. In recent times, Amtrak owned a substantial number of out-of-service cars, which it could not afford to repair. Funding repair gives the passenger rail agency assured availability of cars for a pro- posed service at reduced cost. Exact terms of an agreement have to be worked out with Amtrak and would depend on the details of the rehabilitation project and the expected remaining service life of the cars. – Use Amtrak-provided cars, the cost of which will be included with Amtrak’s charges for train operations and equipment maintenance for the proposed service. In recent years, Amtrak has lacked funds to purchase new cars or to rehabilitate more than a minimum number of out-of-service cars. This meant Amtrak was unable to provide cars for new intercity services, and passenger rail agencies had to adopt one of the previously mentioned alternatives. Since PRIIA (2008) and ARRA (2009), however, Amtrak is pressing ahead with plans to specify and acquire new cars both to replace life-expired equipment and build its fleet to support new services. • Commuter Service – Purchase new equipment as previously described for intercity service. – Lease new equipment as previously described for intercity service. – Acquire existing commuter equipment suitable for the proposed service, by outright pur- chase, long-term capital lease, or short-term lease of temporarily surplus cars from another commuter rail agency. An example of a short-term lease was the temporary use of Sound Transit Bombardier cars by VRE and Los Angeles Metrolink until increases in demand and service frequency in Seattle required their return. As well as selecting equipment for a specific service and developing a process for obtaining suitable equipment, the passenger rail agency has to determine the number of cars required. Fleet planning can be a simple manual process for smaller passenger services, usually involv- ing calculating trip and terminal turn-around times and designing a schedule for a few train sets. For example, the New England Downeaster service is operated with only two train sets. For more complex operations, a computer simulation can help optimize fleet operations. For example, SYSTRA’s RAILSIM package has a module that provides this capability. There can be substantial cost leverage in good fleet planning. For example, it was possible to add service frequencies on California’s Capitol Corridor without adding train sets by reducing layover and turn-around times and rescheduling some trips. Similarly, the proposals to extend the Downeaster service to Brunswick, Maine, can be managed without adding train sets by using layover time in Portland, Maine. 3.3.5 Signaling and Train Control Capital Costs Issues related to train control systems should be resolved in capital planning and, in the future, will inevitably be centered around meeting PTC requirements in the RSIA. New advanced tech- nology PTC systems have been an area of research interest in the railroad industry for over 20 years. The primary function of PTC is to enforce safe speeds at all times to prevent collisions, over- speed events, travel over misaligned switches, and intrusion into work zones. More advanced versions of PTC may also provide capacity benefits from moving block train spacing, more pre- cise speed control, and location management. 44 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

The RSIA requires PTC installation by 2015 on most freight main lines used for passenger service and/or for conveying certain hazardous materials. The same legislation required the rail- road industry, in consultation with the FRA, to develop PTC interoperability standards. Since then, both the railroad industry and the FRA have been actively making plans to implement the legislation, but many of the details are still to be determined. Further discussion of the issues relating to PTC implementation since the October 2008 legislation and FRA safety regulations are provided in Appendices B and C. Even though the details of PTC as applied to the U.S. rail system are still to be worked out, passenger rail operators outside the NEC area will clearly be affected: • With few exceptions, locomotives and multiple-unit driving cabs will have to be equipped with PTC equipment that conforms to national interoperability standards. Each passenger service operator will be responsible for the costs of equipping its own rolling stock. • Passenger rail agencies may have to bear all or part of the costs of wayside PTC apparatus on routes used by their trains. The actual shares will depend on whether the PTC is needed any- way because of corridor use for specified hazardous materials, corridor ownership, and whether the passenger service is Amtrak intercity or commuter. Cost-sharing arrangements will have to be negotiated among users, and it is possible that the STB will become involved in resolv- ing disputes in this area and establishing cost-sharing principles. It is not clear at this point whether a “basic” PTC system installed in response to RSIA require- ments for a hazardous materials corridor will also allow passenger trains to operate at speeds exceeding 79 mph, or whether additional requirements will be imposed above 79 mph or at higher speed thresholds. Passenger rail interests need to be engaged with the FRA as require- ments for higher-speed operations are developed. The uncertainties include what additional requirements, if any, may be imposed on freight operations sharing track with passenger oper- ations exceeding 79 mph. 3.4 Operations and Maintenance Costs and Cost Sharing 3.4.1 Overview This section discusses the technical issues associated with estimating how O&M costs should be distributed among each railroad corridor user in a corridor shared by multiple users, includ- ing the differences between avoidable and fully allocated costs. The different cost categories are identified, together with the operating parameters that drive costs. This is followed by a discus- sion of methods used to estimate costs as a function of usage by freight and passenger trains. The discussion is primarily concerned with the costs of operating a privately owned freight railroad that hosts a passenger rail service. Costing principles are the same where the host railroad is owned by a public agency, but the public agency normally does not expect to make a profit on invested capital and may treat expenses like depreciation differently. The resulting cost estimates may be used to support contractual agreements between host and tenant as described in Chapter 4 and the process for amending and updating contracts as described in Chapter 5. Much of the material in this section is taken from a predecessor study, “Cost Allocation Methods for Commuter, Intercity and Freight Rail Operations on Shared-Use Rail Systems and Corridors” (AECOM 2006). Allocating railroad infrastructure and overhead costs fairly among different types of rail traf- fic is technically difficult, and in many situations there is really no single “right” answer. Once a rail line is put in place, a significant fraction of infrastructure O&M costs are fixed and can only be changed by either changing track quality or adding or removing features of the route, such as Analysis and Modeling 45

a passing siding. Allocation of fixed costs will always be somewhat arbitrary. In addition, many infrastructure components subject to wear or degradation, such as rails and ties, have long lives and annual costs will vary over time as the components age. A decision has to be made between sharing costs “as incurred,” recognizing that there could be considerable year-to-year variation as components age and when substantial renewals become due, or estimating long-run average costs and charging the same amount each year. A further complexity is that a passenger rail agency might fund a track renewal, as on the Downeaster route in Maine, and should enjoy the benefit of reduced annual maintenance costs in following years, instead of contributing to the next renewal cycle. Because of the technical complexity and the presence of substantial fixed costs, there is a long (and contentious) history of methods to allocate railroad operating costs to a particular traffic. In the regulated era prior to the 1980 Staggers Act, much of the effort was directed at determining “fair” freight rates to charge freight shippers of different commodities. These allo- cation methods relied on historical cost data that the railroads were required to submit to the ICC. Statistical analysis of the data yielded a costing formula, which was then used to estimate the cost of a specific freight movement. Similar methods are still used by the STB to support the more limited freight rate regulation in force today. The STB’s responsibilities for regulat- ing passenger fares was eliminated with the formation of Amtrak, but it still has the responsi- bility to mediate disputes between Amtrak and freight railroads regarding access charges. Appendix B contains a discussion of the various costing issues brought before the STB for res- olution during negotiations between Amtrak and the host railroad for the Downeaster service, and the resulting decisions. These are the most recent series of decisions by the STB on Amtrak incremental costing. The following sections first introduce the primary cost categories that make up O&M costs, including periodic like-for-like replacements of worn-out track components such as rail. Then, the definitions and differences between avoidable and fully allocated costs are dis- cussed, including how cost estimates may be used in calculating the share of costs allocated to freight and passenger operations using the same tracks. Finally, methods to estimate costs of a specific rail operation are discussed, including using historic actual costs and engineer- ing analysis. 3.4.2 Railroad Operations and Maintenance Cost Categories Table 3-1 shows the matrix of railroad O&M cost categories organized by operating function and by the nature of the expense. The shaded blocks in the table matrix highlight the functional and expense categories that may need to be apportioned among the users where a rail line is shared by multiple services. The rationale for sharing the costs among users within each of the expense categories is discussed in the following paragraphs. Transportation Labor Because these are not shared resources, in most cases, each operator employs its own train crews and other on-board personnel. In some cases, a freight railroad may provide passenger train crews (as on some Chicago area commuter lines and in Seattle), but this is usually the subject of a separate operating services contract. In contrast, dispatchers are always a shared resource, because one dispatch desk (workstation) has to be responsible for all trains on a rail line segment. Therefore, dispatching costs are always a line item in a shared cost analysis and would be shared between users per applicable agreements. For Amtrak intercity services to which Amtrak incremental costing applies, only incremental dispatching costs, if any, will be charged to Amtrak. Where two passenger operators share the same rail line, it may also be necessary to share certain station costs. 46 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

Transportation Fuel and Power In most cases, each operator uses its own fleet of locomotives, and diesel fuel is purchased sep- arately by or on behalf of each operator. On a shared electrified railroad, power costs must be shared. This calculation can be complex, because there is no easy way to directly meter the power consumed by each train. Equipment Inspection and Maintenance Most passenger operators provide and maintain the equipment (cars and locomotives) that is dedicated to their services. Occasionally, one agency will rent or lease cars to another agency, but any associated maintenance services provided are the subject of a separate specific agreement. Therefore, cost-sharing issues rarely arise. Maintenance-of-Way and Structures Maintenance-of-way functions are always under the control of the host railroad. This is the most complex area related to cost sharing, because some cost categories are largely fixed (i.e., do not vary with use) and other costs vary with traffic volume and types of use in different ways. Also to be resolved is the approach to costs for periodic replacement of wearing components, such as rail. A rail replacement project may be treated as a capital cost in accounting terms, to be depreciated over the life of the rail, but the consumption of rail life may be treated as a mainte- nance cost in a cost allocation calculation. Most efforts to find realistic costing and allocation methods are concentrated in the maintenance-of-way and structures area. Maintenance of sig- nal and telecommunications systems also fall into this area and are mostly fixed costs that vary little with the level of usage once specific systems are installed. Analysis and Modeling 47 Operating Function Labor Materials and Supplies Fuel and Power Other Expenses Transportation Line-haul train crews, Switching and yard crews, On-board services (passenger), Passenger station staffing (passenger), Dispatchers, Supervisors Minor items Electric power for traction on electrified lines, Diesel fuel, etc. Equipment Maintenance Inspectors, Shop and terminal maintenance employees Spare parts and materials for car and locomotive maintenance Power for workshops and power tools Contracted maintenance services, e.g., rebuilding components, Depreciation Way and Structures Maintenance Track inspection and maintenance, Signal and telecoms, Inspection and maintenance, Structures inspection and maintenance Rails, Ties, Ballast, Replacement parts for signal and telecoms, Other construction materials Diesel fuel for work trains and maintenance equipment Equipment rental, Contract services, Depreciation General and Administrative (G&A) Management above first- line supervisor and technical specialists, Administrative staff of all types, Accounting, Legal services Office supplies Power for building services Office rentals or leases, Contract services, e.g., for buildings, Insurance, Loss and Damage payments Table 3-1. Operations and maintenance cost categories.

General and Administrative Activities These are functions of the host (usually freight) railroad, which will devote some fraction of its effort to supporting tenant passenger operations. Some expenses can be directly attributed to the passenger operation, such as staff dedicated to managing the interfaces with the passenger operators, or specific insurances only required with passenger service. However, most adminis- trative functions cannot be allocated in this way, and an alternative approach must be sought. Examples might be an allocation based on an overhead percentage, an allocation based on train- miles or some other measure of output, or a mix of both approaches. 3.4.3 Cost-Sharing and Allocation Approaches Before methods to estimate costs and the appropriate distribution of costs between multiple rail corridor users can be discussed, it is helpful to define the two principal costing concepts that are used when estimating cost sharing between host and tenant. These concepts are avoidable costs and fully allocated costs. Avoidable Costs Avoidable costs (sometimes called incremental or marginal costs) are those short- and medium-term costs that would be directly avoided if the tenant’s operations ceased. For a pas- senger rail tenant operating on a freight railroad, avoidable costs typically include wear and dete- rioration of track and structures that can be attributed to the passenger operation, a share of dispatcher costs, and the cost of management directly associated with the passenger service. Fixed costs of operating the railroad and compensation for utilizing an increment of capacity are not included. This approach is used to compensate freight railroads for accommodating Amtrak services, as required by the legislation that established Amtrak as the national intercity passen- ger rail operator. Estimating avoidable cost becomes more complex if track quality or signal system improve- ments are required for a new passenger service. In this case, the avoidable-cost principle would require the passenger operator to be responsible for the difference between maintaining the higher quality under combined freight and passenger operations and what it would have cost the freight railroad for maintenance to freight service requirements. If the passenger agency had funded capital improvements to track and signals, then there is a further complication: mainte- nance costs to the freight railroad would be reduced for a number of years, until track compo- nents became due for maintenance or replacement. This issue was one of those at dispute between Amtrak and the host railroad prior to implementation of the Downeaster service, as dis- cussed in Appendix B, Section B.2.2. At present Amtrak is the designated operator of all intercity passenger rail services on shared corridors, using Amtrak’s rights of access at avoidable costs. Amtrak already has operating agree- ments with all the major railroads, which include agreed-upon compensation for track use, usu- ally expressed as a cost per train-mile. Amtrak and the host railroads will negotiate an extension of this agreement for a new service with an amended track-use fee, where appropriate, to reflect local conditions, including infrastructure investments funded by the passenger agency. Any dis- putes between Amtrak and the host railroad regarding the avoidable-cost fees for a specific ser- vice can be referred to the STB or the NAP for resolution. Under most Amtrak agreements with freight railroads, the NAP would resolve disputes over the application of the costing methodol- ogy in that agreement to a new service. Fully Allocated Costs In this approach, the host railroad’s full costs are shared between the host and tenant on a log- ical basis. A freight railroad may also add a charge to the allocated full O&M cost to provide a 48 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

return on the capital investment in rail corridor capacity used by a passenger service, if the pas- senger agency has not used one of the alternative methods of paying for the capital cost of capac- ity as discussed is Section 3.3.1. Essentially, the freight railroad treats the passenger agency as it would a freight customer. A railroad may also agree to a hybrid approach where a share of some, but not all, fixed and overhead costs are allocated to the passenger operation, but this scenario is likely only if the agreement is part of a broader agreement that includes some other benefit for the freight railroad. Fully allocated costing is most commonly used for commuter service over freight railroad tracks, where Amtrak access rights do not apply. The primary issue is establishing a full cost allo- cation method that is logical and acceptable to both parties. Establishing such a method can be straightforward where two freight railroads share the same tracks, because the types and quality of service required by host and tenant are the same. A freight railroad–to–freight railroad track- age rights agreement will typically apportion costs in proportion to car-miles operated over the shared territory. Where the tenant is a passenger service, arriving at a fair and logical cost division is more challenging, both at the technical level in deciding how costs are influenced by the pres- ence of passenger service and in reaching a mutually acceptable formula in access negotiations. 3.4.4 Operations and Maintenance Cost Analysis This section addresses two interlocking areas: the technical basis for determining the share of each cost category allocated to each service and the approaches used to incorporate technical understanding into shared track agreements for intercity and commuter services. This section is not intended to cover the specific cost allocations used for Amtrak avoidable-cost agreements based on the Amtrak statute and STC decisions, but rather to describe the technical and logical factors that drive costs on a shared corridor. Relevant cost categories include: • Dispatching. Dispatcher workload is driven by the number of track miles in the dispatcher’s territory, train-miles operated, and the complexity of operations. Adding passenger trains on a freight corridor is likely to add to dispatching complexity, given speed differences and sta- tion stops, adding workload. A logical approach for dispatching costs might be to allocate according to train-miles with an additional weighting for passenger trains. In some cases, a passenger operator has agreed to fund an additional dispatcher position so that passenger ser- vice can receive more focused attention, potentially reducing delays. Alternatively the passen- ger rail agency may fund a coordinator as a liaison between the host railroad and the passenger operator and agency, to help manage passenger service performance. • Electric Power. Because there are no current electrified freight operations in the United States, this is an issue that arises when two electric passenger operations share tracks, as in the NEC. Because it is not easy to meter power at the point of use on each train, the best approach may be to estimate per-mile power consumption as a function of train size (number of cars) and type of service, using the results of TPC analysis and electric power prices to establish and cal- ibrate a per-train-mile charge. • Track Maintenance. This area is the most complex for cost allocation, because track degra- dation under different traffic types is difficult to quantify. The most comprehensive studies on this issue have been by Zeta-Tech, who developed the TrackShare model, described in papers and presentations for the Transportation Research Board and in a report prepared for the FRA (Zeta-Tech 2004). This method had been used in a number of settings, including cost estimat- ing for the proposed Midwest network, cost allocations between users of the NEC, and cost allocation for Amtrak service on freight railroads. Zeta-Tech takes a fundamental engineering approach, calculating the cost of wear and degradation of track and track components as a func- tion of loads on the track. These costs are a function of axleload, speed, and track characteris- tics such as FRA track class, grade, and curvature. The model yields Engineering Adjustment Analysis and Modeling 49

Factors (EAFs) to apply to a base cost estimate as a function of traffic type and track characteris- tics to obtain a cost estimate for a specific traffic mix. This model is particularly useful for calcu- lating a logical division of costs when a passenger tenant requires a higher track quality than the host freight railroad. Because the model is too complex to simply incorporate into a cost-sharing contract, the usual process is to use the analysis to estimate a per-car-mile or per-train-mile charge for the planned service and incorporate the result into the agreement. • Track Inspection. Both FRA regulations and industry practice require regular track inspections for defects. The presence of passenger service may or may not trigger a change in inspection practice depending on specific traffic levels and speeds. If a change in practice is required, then the additional cost would be allocated to the passenger expense. Otherwise, costs would be shared among the services in a fully allocated approach, most logically as a function of ton-miles adjusted by EAFs. • Other Infrastructure Inspection and Maintenance. This area covers bridges and structures as well as signal and train control equipment. Most of these expenses are only weakly dependent on traffic level and type, and cost allocation is somewhat arbitrary. An allocation based on adjusted ton-miles for structures and train-miles for signal and train control systems would be logical. Some passenger services may require improved signal and train control systems to permit increased speeds or to add capacity. If this is the case, all or part of the additional cost of inspecting and maintaining the improved systems over the original installation would be the responsibility of the passenger operator. For example a passenger operator and host railroad may fund a conversion of automatic block signaling to centralized traffic control and add pow- ered switches at passing sidings. Because the improvements benefit both freight and passenger service, both host and tenant would share additional inspection and maintenance costs, if any. • General and Administrative Expenses. In most cases, there is no specific cause-and-effect relationship between selected G&A expenses and the level and mix of train operations. The only exception is where staff are specifically dedicated to the passenger affairs. Otherwise, the allocation may be determined in proportion to train-miles or represented as a percentage or fee on other passenger-related expenses, or, in some cases, the parties may agree to a fixed annual fee. 3.4.5 Application to Intercity and Commuter Operations The foregoing discussion summarizes the technical issues involved in cost sharing. The man- ner in which these technical issues are brought into negotiations between the passenger rail agency and a host freight railroad depends on the details of each service, as summarized in the follow- ing paragraphs. Amtrak Intercity Service with No Service-Specific Infrastructure Investments Amtrak compensates host freight railroads for intercity passenger operations on an avoidable- cost basis. Agreements for a new or expanded service will be based on existing Amtrak operat- ing agreements with mutually agreed-upon variations to reflect specific local circumstances. Disputes can be resolved by the NAP or by the STB, which has the power to impose a decision on Amtrak and the host railroad. Amtrak Intercity Service where a State Passenger Rail Agency Has Funded Added Rail Corridor Capacity or Upgraded Infrastructure Amtrak will negotiate a track-use fee agreement with the host railroad for the specific service with the host railroad, proposing fees based on previous experience in comparable situations, and taking into account the passenger agency’s investment, track class, and added signal and train con- trol equipment. The parties are free to support their position with technical analyses, but there is no requirement for such analysis. If Amtrak and the host railroad cannot agree, then they can 50 Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors

appeal to the STB, supporting their position as they see fit. The state or regional passenger rail agency must work with Amtrak and may choose to directly fund a track maintenance gang or a dedicated dispatcher to ensure service quality. Commuter Operations Hosted by a Freight Railroad The commuter rail agency will have to compensate the freight railroad for all or most fully allo- cated costs. The details of individual agreements are highly variable and will reflect local circum- stances, especially track and signal system improvements funded by the commuter rail agency. Many of the sharing approaches mentioned previously regarding individual O&M expense items and technical cost analysis may be reflected in the resulting agreement. However, it is essential that the final agreement is simple to administer, such as a formula that uses easily determined operations metrics, such as car- and locomotive-miles and train-miles. Analysis and Modeling 51

Next: Chapter 4 - Content of Shared-Use Access and Operating Agreements »
Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 657: Guidebook for Implementing Passenger Rail Service on Shared Passenger and Freight Corridors is designed to aid states in developing public–private partnerships with private freight railroads to permit operation of passenger services over shared-use rail corridors.

The guidebook explores improved principles, processes, and methods to support agreements on access, allocation of operation and maintenance costs, capacity allocation, operational issues, future responsibilities for infrastructure improvements, and other fundamental issues.

  1. ×

    Welcome to OpenBook!

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

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

    No Thanks Take a Tour »
  2. ×

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

    « Back Next »
  3. ×

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

    « Back Next »
  4. ×

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

    « Back Next »
  5. ×

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

    « Back Next »
  6. ×

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

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

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

    « Back Next »
Stay Connected!