Technical Document and User Guide for the Multi-Modal Passenger Simulation Model for Comparing Passenger Rail Energy Consumption with Competing Modes (2015)

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.

iv Summary The objective of NCRRP Project 02-01, “Comparison of Passenger Rail Energy Consumption with Competing Modes,” is to provide like-for-like comparisons of energy consumption and greenhouse gas (GHG) emissions of commuter and intercity passenger rail operations with competing modes of travel. In the context of this research, “passenger rail” includes higher speed, high speed, intercity, and commuter rail operations - those rail systems that are operated under the jurisdiction of the Federal Railroad Administration (FRA). “Competing modes of travel” include passenger automobiles, light-duty trucks used for personal transportation, suburban commuter bus services, intercity bus services, and air transportation. This research project involved a literature review to collect data on relevant passenger transportation system characteristics and performance, formulation of an analytical framework for each passenger transportation mode considered, development of a quantitative decision-support tool based on these analytical frameworks, and finally applying the decision-support tool to numerous case studies which explore a range of commuter and intercity passenger rail operations and compares their fuel/energy consumptions and GHG emissions with those of competing modes of transportation for comparable door-to-door trips. The quantitative decision-support tool developed specifically for this research is called the Multi-Modal Passenger Simulator (MMPASSIM). It is an open-source Microsoft Excel macro enabled worksheet which may be used to predict and compare fuel and energy consumption and GHG emissions for a wide range of commuter and intercity passenger transportation scenarios. The results of this research project are documented in two parts: NCRRP Report 3: Comparison of Passenger Rail Energy Consumption with Competing Modes and NCRRP Web-Only Document 1: Technical Document and User Guide for the Multi- Modal Passenger Simulation Model for Comparing Passenger Rail Energy Consumption with Competing Modes. The purpose of this NCRRP WOD 1 is to describe the technical details of the analytical framework underlying the MMPASSIM quantitative decision- support tool and also provide documentation and guidance on how to set up and use the MMPASSIM model to evaluate passenger transportation alternatives. NCRRP Report 3 documents the findings of the literature review and details a series of rail technology evaluation and modal comparison case studies performed in this research project. The MMPASSIM spreadsheet files, together with the Technical Document and User Guide, are provided on CRP-CD 176, which accompanies NCRRP Report 3. The MMPASSIM files also may be downloaded from the NCRRP Report 3 web page at www.trb.org. The analytic framework developed uses common energy and emissions metrics across transportation modes and fuel types in order to facilitate like-for-like comparisons. Each modal trip involves a principal leg and, optionally, up to five access and egress legs. However, the focus is on the principal modal leg of a full door-to-door trip which requires relatively detailed simulations. The calculated energy and emissions performance over a principal leg will differ from simpler average performance measures of a transportation mode. The energy and emissions performance for access and egress legs are assessed using default average performance metrics provided for each of the access and egress modes selected. Each transportation mode is modeled, within the limits of publicly available data, such that seasonal, regional and equipment-specific characteristics are reflected in the modal energy and emissions performance. Examples of the limitations of publicly available data in the rail mode are auxiliary power requirements for rail, resistance coefficients for specific types of rail equipment, and engine performance for recent vintage locomotives.

v Intercity scheduled bus operations also do not have load factors by region or service. This document and the case study analysis in the companion document indicate where estimates were used to characterize modal characteristics. A common element of the mode-specific simulation modules used in the spreadsheet-based model is the ability to specify the equipment and route characteristics involved in making the specific trip of interest. Default choices are available for the user to select in a quick simulation comparison; however, the user also has the ability to modify the default characteristics or define new characteristics for each transportation mode if desired. Railway Mode Model The passenger train simulation implemented in the MMPASSIM model, unlike a traditional Train Performance Calculator (TPC), simplifies energy and emission calculations by treating passenger trains as a lumped mass. This approach is feasible since passenger train consists are short and light with high power to weight ratios such that their performance is not limited by track grades. However, despite this simplification, the influence of gradient and train length on the calculated energy and GHG emission intensities are included in the model. A rail trip is specified by defining the characteristics of a passenger train and the route characteristics of the track over which it is to operate. The train characterization includes: the length and masses of the vehicles, the passenger capacity and load factor, parameters defining the inherent train resistance, and the auxiliary load and traction power capabilities. Inherent train resistance is modeled using a quadratic relationship which provides for a constant term to represent rolling friction and ground hysteresis losses, a speed-sensitive term representing dynamic rolling losses, and a term associated with aerodynamic losses which varies with the square of train speed. The variation of a locomotive or power car’s maximum tractive effort with train speed is modeled using a multi-segmented relationship where each segment of the tractive effort versus speed curve is described using an equation with a constant term, a linear speed- sensitive term and a term which varies inversely with train speed. These equation coefficients used in combination with up to a maximum of 5 speed segments facilitates accurate modeling of diesel-electric locomotives and electric power cars. The tractive effort for most diesel-electric locomotives may be accurately characterized using a 2-segment curve where the first segment represents a linear low-speed torque limited region and the second segment represents a power limited region where the tractive effort decreases inversely proportional to speed. Modeling electric power car tractive effort often requires additional straight line segments between the low speed tractive effort limit and a high-speed power limited region, and the highest speed region may also exhibit a fall-off in tractive effort beyond that of a constant power relationship. The additional limitation on passenger locomotive acceleration performance due to diesel engine loading rate is also modeled. The default rail vehicle data set included with the MMPASSIM model includes characterizations for a number of commuter and intercity passenger rail consist. These may be used directly, modified by a user or used as templates to develop new passenger rail consist. The rail model’s route characterization is based on data that will usually be available to a rail-agency and rail system operators but may not be accessible to the general public. One of the key influences of passenger train performance is the number of speed changes involved on a route. Permanent speed limits shown in railway timetables are more generally available than are track gradient profiles and we recommend that actual speed limit tables be used in simulating a passenger rail service wherever possible. These are input into the model by specifying the start location of all speed limit changes on the route. The location of all stops to be made along a route must also be specified.

vi Gradients have less impact on passenger train performance and for this reason the model is able to use a more generalized gradient distribution representation instead of the detailed grade profiles specified in a railway track chart. The gradient distribution used in the rail model summarizes the actual track grades in a segment of track into six gradient bins which categorize the actual average of all grades collected in a gradient bin and the percent of total track segment distance associated with all grades in each bin. A route may be characterized using up to eight separate segments having different grade distributions when there are significant differences in track profile over the route’s overall length. The model is provided with several default region and service-specific rail route grade characteristics which were developed and used for case studies. These grade characteristics were developed using a range of actual track gradient profiles and may be used with reasonable accuracy when grade profiles for a specific route cannot be supplied. The rail mode simulation uses the passenger equipment’s inherent resistance and tractive effort characteristics to calculate detailed profiles of acceleration, coasting and braking performance which can be achieved on level and tangent track. These calculations are stored in one mile per hour increments in lookup tables which the model then uses as the basis for determining the duration of acceleration and braking operations required at all speed limit changes and for stops. This provides for a more computationally efficient method of evaluating passenger train performance than a second-by-second simulation of movement along the entire track. The MMPASSIM model performs a rail mode simulation by stepping through a route and evaluating the duration of acceleration, cruising (at a constant speed limit), and braking phases for each section of track bounded by a speed limit change. The work done at the rails to overcome inherent resistance during cruise and acceleration phases is broken out into rolling, dynamic and aerodynamic components and the additional work done by brakes to maintain speed limits while traveling on downgrades or for speed reductions and stops is calculated. The rail module accumulates the energy associated with each of the sub- categories of energy dissipation such that the effectiveness of alternative technologies can be gauged from the model output for a single train run. Highway Mode (LDV and Bus) Models The highway modes’ energy and emissions performance in MMPASSIM are derived by simulating the second-by-second movement of a specified vehicle over a set of pre- defined speed profiles (drive schedules) in urban areas and uses the vehicle’s cruise performance for calculations over rural intercity segments. Vehicle resistance to motion is based on rolling and aerodynamic resistance coefficients, vehicle power and energy performance is governed by specified engine and drivetrain characteristics and auxiliary loads are associated with regional climates as specified for three seasons of the year (summer, winter, and other). Eight drive schedules are implemented in the model to characterize the influence of traffic congestion on the movement of a simulated LDV or bus in urban areas. These were selected from a range of speed profiles developed by the U.S. Environmental Protection Agency (EPA). The specific congestion performance for a simulation is characterized by specifying the proportion of a trip which is to follow the speed profile of each of these eight drive schedules. These proportions may be individually specified for 5 different time-of-day congestion intervals to allow for traffic variability and default sets of values are provided to represent driving through typical large and small cities in these 5 time periods. The time-of-day periods include: • a.m. peak • p.m. peak • mid-day • shoulder periods (next to peak periods)

vii • overnight Users of the MMPASSIM model are also able to develop customized sets of time-of-day based drive schedule proportions specifically tailored to urban areas of interest. Note that the light duty vehicle (LDV) drive schedules contain a high-acceleration performance drive schedule that is not included in the bus drive schedules; however, the simulation process is the same for both modules. Default bus performance characteristics are provided for a limited set of four representative bus types as follows: • 45 foot bus with 56 passenger seats • 41 foot bus with 48 passenger seats • double deck bus with 81 passenger seats • hybrid commuter bus with 57 seats These typical bus types are characterized using publicly available data for resistance coefficients, drivetrain efficiency, auxiliary loads and diesel engine efficiency and do not represent any particular manufacturer’s vehicle. Commuter bus operating characteristics and energy intensity performance are calibrated using data reported for commuter buses by municipal operators in the Federal Transit Administration’s (FTA) National Transit Database (NTD). The intercity buses are calibrated using data reported in the NTD for short-distance intercity commuter bus travel combined with data reported in a 2013 survey of major North American bus operators undertaken for the American Bus Association. The average fuel economy for the two sources is 42.3 L/100-km or 5.59 mpg (average of 6.09 and 5.08 mpg from the ABA and NTD sources respectively). The default light duty vehicle performance characteristics provided in this model were derived from sales-weighted class-average values published annually by the U.S. Environmental Protection Agency (EPA). Specifically, the characteristics of 2011 model- year LDVs were grouped on the basis of similar coast-down resistance coefficients to develop a set of six representative vehicle classes as follows: • Small cars (Sm-Car) • Midsize cars and all station wagons (Mid-car/SW) • Minivans and non-truck SUVs (Mini-V/sm-tSUV) • Large cars, medium-truck SUVs and small pickup trucks (Lg-car/cSUV/smPU) • Large pickup trucks (PU) • Large truck SUVs (Lg-tSUV) The six individual class-average vehicles are provided as user-selectable default vehicles in the MMPASSIM model. In addition, three composite vehicles are provided which typify the performance of the mix of vehicles used for specific types of trips which include: • a composite based on the estimated mix of personal LDVs used in local trips • a composite based on the estimated mix of personal LDVs used for intercity trips • a composite based on the estimated LDV mix for taxis Additional “sales-weighted” and “driven-fleet” composite vehicles were developed from EPA fuel economy data for the 2011 model year. These class-average performance characteristics were developed by applying class-specific modifiers to a generic engine fuel- map and a generic six-speed transmission such that the EPA’s sales-weighted average performance is obtained when simulating operation over the EPA’s underlying certification drive schedules. A similar process is used to characterize the 2011 “driven fleet” to reflect the relative performance of older vehicles with an appropriate vehicle age distribution. An algorithm is provided to create “sales-weighted” and “driven-fleet” composite LDV

viii characterizations for years beyond the 2011 base data year. Default composite values for “sales-weighted” and “driven-fleets” for 2012 and 2013 are provided in MMPASSIM and the model architecture supports addition of future year composite fleet values as EPA estimates of fuel economy for those years become available. The EPA estimates that actual driving conditions lead to fuel economies which fall below the reported values by between 12% and 15% due to various factors not considered in the 5 cycle test process. These include: • Road and tire condition • Effect of wind on aerodynamic drag • Effect of temperature on aerodynamic drag • Seasonal cold-start fuel consumption • Seasonal auxiliary power loads • Road grade effects The MMPASSIM model considers the impact which these factors have on the running performance of LDVs. However, driver-behaviour, auxiliary power usage and vehicle maintenance can all affect the fuel economy of a specific vehicle and there will always be a range of fuel economy performance around any derived average. Our objective is to inherently model most parameters and provide an average in-service fuel economy that is representative of the real-world experience. Air Mode Model Air mode trips are analyzed differently than rail and highway mode trips. The flight of a specific aircraft is not simulated on a second-by-second basis while accumulating distance traveled and fuel consumed in moving from the origin to the destination. Rather, the fuel consumption and emissions intensities for an air trip are derived by applying fuel intensity coefficients to the distance traveled. This approach is feasible since very detailed aircraft in- service performance data is published by the U.S. Bureau of Transportation Statistics (BTS). The route to be followed is defined by specifying a sequence of up to four IATA (International Air Transport Association) airport codes. A direct flight would involve only two IATA airport codes, one for the departure airport and the other for the destination airport, while multi-leg flights may be configured with up to three legs by specifying up to two intermediate airport codes. The air-mode model automatically computes the great- circle distance (shortest distance along the earth’s sphere) traveled between two airports based on their latitude and longitude. A user may easily expand the list of available airports by adding IATA codes, latitude and longitude. The default aircraft characterization data provided with the model are based on 2011- 2012 operations of domestic U.S. scheduled air carriers and can be updated by the user as desired in future years as air technology and operations practices change. The default data is organized into the following five categories of aircraft: 1. Turboprops (TP) 2. Small Regional Jets (SRJ) (defined here as jet aircraft with less than 50 seats) 3. Regional Jets (RJ) (defined here as short-range jet aircraft with 50 to 89 seats) 4. Narrow Body Jets (NBJ) (defined here as jet aircraft with greater than 89 seats in a single aisle configuration) 5. Wide Body Jets (WBJ) (defined here as jet aircraft with greater than 89 seats configured with more than one aisle) Air mode trips are assumed to be serviced by a distribution of these five aircraft types rather

ix than by a specific aircraft. A default mix of aircraft types was derived from the BTS data and is provided for seven ranges of trip length. The air model automatically assigns the proportions of aircraft types used in each leg of an air trip according to the distributions associated with the applicable trip length range. However, a user may adjust the proportions of aircraft types assigned to each leg of an air trip. Fuel intensity coefficients for each representative aircraft category have been derived separately for the LTO (landing and takeoff) and cruise phases of a flight. The cruise phase fuel intensity values have been adjusted to account for the average incremental distance traveled in excess of the shortest path between origin and destination under both peak and off-peak period conditions. The fuel consumed in the cruise phase of an air trip leg is computed by multiplying the great-circle distance traveled by the fuel intensity coefficient for the representative aircraft type. Where a distribution of aircraft types is specified, the calculation applies the fuel intensities in those proportions. Access and Egress Modes Only the primary transportation modes being compared are simulated in detail in the model whereas the performance attributes of access and egress modes are simple averages provided in default lookup tables. The model architecture supports specification of different performance attributes for an access/egress mode to be associated with city size, time-of- day, day-of-week and season. The attributes of public transportation modes have been derived from the 2011 National Transit Database’s Service and Energy Tables. The electricity supply used for public transportation modes is region-dependant but all other performance metrics are based on one average applied to all regions. Attributes for personal automobiles and taxis were derived with the detailed LDV simulation model in a one-time simulation of the 2011 driven-fleet composite vehicle. The following assumptions were made for the highway access/egress modes: • taxis were assumed to travel 1.5 km for every km of passenger carrying travel • drop-off and pick-up was assumed to have 60% return-to-origin travel and 40% being part of a 2-person trip that incurs 10% extra travel distance • carpools are assumed to involve 3 persons and the trip length is 15% longer than any one-person trip The fuel intensity of highway modes is adapted to congestion conditions via peak and off- peak multipliers, which are specified for three city sizes (large, small and rural municipality). Regional Influences The MMPASSIM model provides a default regional characterization for Northeast, South, Midwest and West regions of the continental U.S. and can be optionally updated by a user to include additional user defined regions. The primary regional influences are seasonal and include travel variations, temperature variations, and average use of climate controls and auxiliary power for ground transportation modes. Congestion factors for travel in urban and rural areas during peak and off-peak periods are also specified for a region. Finally, the fuel and emissions intensities of access and egress modes are defined for each region. Electricity generation is further disaggregated into nine sub-regions with the distribution of fuels used in generating that electricity derived from U.S. Energy Information Administration data for 2011 and upstream fuels consumed were derived from the GREET model. MMPASSIM Model Structure The MMPASSIM model performs multi-modal door-to-door passenger trip comparisons of energy and GHG intensity. Additional details on energy dissipation sources are provided in conjunction with rail-only simulations to facilitate comparisons between different rail technologies. The model is implemented as a macro enabled Microsoft Excel worksheet which supports the following three types of analyses:

x 1. Single Train Simulation - a single train service is assessed for its performance and energy/GHG emissions breakout. 2. Rail Technology Evaluation - a comparison of up to four passenger rail technologies to compare and assess the energy/GHG emissions savings realized. 3. Mode Comparison - a comparison of up to four passenger modes (rail, bus, air, light duty vehicle) to compare and assess the energy/GHG performance in a door-to-door trip. The primary user interface for all simulations is provided in the ‘Master-I-O’ worksheet (blue tab) from where a user configures all required simulations for an analysis. However each transportation mode is configured and operated as a semi-independent sub-model which may be configured and controlled independently by a user if desired. The sub-model user interfaces are provided in the ‘Rail-I-O’, ‘Air-I-O’, ‘Bus-I-O’ and ‘LDV-I-O’ worksheets (all with blue tabs). A system of pop-up user forms (menus) and Visual Basic for Applications (VBA) macros coordinates the configuration of all desired simulations as well as the transfer of data to and from the sub-model worksheets. The results are displayed to the user in formatted output tables on the ‘Master-I-O’ worksheet which are automatically brought into view while the simulation proceeds. Simulations can be specified from the Master-I-O sheet using the wide range of case study defaults for routes and equipment in each mode. In addition new routes and equipment can be specified and default values can be modified within the model’s other worksheets as discussed below. The model uses green to indicate model inputs, yellow to indicate default values, orange to indicate calculation formulae and blue to indicate output sheets/areas. The ‘Regional-Properties’ worksheet (yellow tab) contains regional data for: average daytime temperature and air conditioning usage by season. In addition it contains default characteristics for fuel, energy and GHG emissions intensities used for the access and egress modes. The access and egress modes support differentiation by size of city and time- of-day congestion for the highway modes. The ‘Energy-Emissions’ worksheet (yellow tab) provides the GHG emissions rates by fuel/energy-source and the indirect (upstream well-to-pump) energy consumption/GHG- emissions associated with each fuel/energy-source. For electricity these factors are also provided for different geographical regions. Mode-specific equipment worksheets (green tabs named ‘Rail-Consist’, ‘Bus-Type’, ‘LDV- Type’ and orange tab named ‘LDV-Resist‘) store the characteristics describing the physical attributes and capabilities of ground transportation mode vehicles. New equipment/vehicles can be introduced into the model by adding data to these worksheets. Mode-specific ‘Route’ worksheets (green tabs named ‘Rail-Route’, ‘Bus-Route’ and ‘LDV- Route’) store the characteristics describing the routes which may be followed by a ground transportation mode. New routes can be introduced into the model by adding to these worksheets. The highway mode drive schedule worksheets (yellow tabs named ‘Bus-Drive-Schedules’ and ‘LDV-Drive-Schedules’) define the second-by-second speed profiles used to represent movement of buses and LDVs in urban areas. The mixes of drive schedules used to represent time-of-day congestion in different urban centers are also declared in those worksheets. Default drive schedule mixes for large and small urban centers have been provided. Users may adjust these defaults and also add new city-specific drive schedule mixes to these worksheets. The ‘Engine’ worksheets (yellow tabs named ‘Bus-Engine’ and ‘LDV-Engine’) provide the needed modal engine efficiency characteristics for the highway modes. Representative fuel

xi maps are used for propulsion systems using non-continuously variable transmissions (applicable to most conventional LDV and buses) while coefficients for a single optimal performance equation are provided for representing vehicles using a continuously variable transmission (CVT). The ‘Simulation’ worksheets (orange tabs named ‘Rail-Simulation’, ‘Bus-Simulation’, ‘LDV- Simulation’ and ‘Air-Simulation’) implement the algorithms used to simulate the movement of a modal vehicle (or a fleet-average characteristic vehicle) representative of the specific service/region being simulated. MMPASSIM Model Outputs The primary MMPASSIM model outputs are provided in simulation mode-specific tables located on the ‘Master-I-O’ worksheet. The output tables are different for each of the three analysis scenarios. The output from a Single Train Simulation is provided in two tables with sufficient detail to permit assessment of the underlying sources of energy consumption and GHG emissions. This information is a first step in validating the input data used for the simulation and in assessing the relative impact that technological changes to specific source components of energy consumption would have. The first table provides the absolute and proportional values of energy consumption and GHG emissions for: seven categories of traction energy (three sub-elements of inherent train resistance, three sub-elements of brake dissipation, and curving resistance); for the traction system’s transmission losses; and for provision of hotel power. The second table provides performance metrics where energy and emissions intensities are output for three divisors (per-trip, per-seat-distance, and per-passenger- distance) and for two service-performance metrics (travel-time and average speed). The output from a Rail Technology Comparison is provided in three tables. The first table provides the same components of energy consumption and GHG emissions as output for a single train simulation but adds additional rows which indicate the percent-reduction in energy and emissions realized by using the alternative technologies. The second table provides performance metrics and indicates the percent reduction compared with the baseline technology case. The third table outputs the total energy and emissions intensities when the indirect consumption and emissions associated with well-to-pump fuel provision are included. The output from a Mode Comparison analysis focuses on performance metrics comparable across transportation modes and expands the comparison to include access and egress legs of a trip. Four tables are output with the same energy/emissions intensity values as were used in the rail technology comparison tables but with an indexed comparison to the baseline rail mode replacing the %-reduction from the baseline rail technology that was used in the technology comparison table. The first table compares direct energy/emission for the modal leg of the trip, the second table compares direct energy/emission for only the access/egress legs of the respective modal trips, the third table compares direct energy/emission for the complete door-to-door trips, and the fourth table compares the full energy/emissions (including indirect well-to-pump) for the complete door-to-door trips. Using the MMPASSIM Model See the front end of the User Guide in Appendix A for an overview and quick reference guide.