June 25, 2009
Regional Transit Planning Tools: A Scan of State of the Art
Next Generation Transit Service Information Portal (TSIP):
Planning White Paper #2
June 25, 2009
Final
by
Nakanishi Research and Consulting, LLC
and
Prepared for:
New York State Department of Transportation
TSIP_Planning_WP2_IndustryScan_Final_v001 Page 5
June 25, 2009
Table of Contents
1.0 Introduction 3
1.1 Objective 3
1.2 Scope of White Paper 3
1.3 Audience 3
2.0 Regional Transit Planning Tools 5
2.1 Overview 5
2.1.1 The Travel Demand Model 7
2.1.2 Current Direction and Future Needs 10
2.2 Transit Planning Tool: T-BEST (Transit Boardings Estimation and Simulation Tool) 12
2.2.1 Introduction 12
2.2.2 Advantages 13
2.2.3 Disadvantages 13
2.2.4 Availability and Cost 13
2.2.5 Vendors 14
2.2.6 Applications 14
2.2.7 Data Requirements 14
2.2.8 GIS Requirements 15
2.3 Puget Sound Regional Council (PSRC) Sketch Planning Tools 16
2.3.1 Introduction 16
2.3.2 Advantages 21
2.3.3 Disadvantages 22
2.3.4 Availability and Cost 22
2.3.5 Vendor 22
2.3.6 Applications 23
2.3.7 Data Requirements 23
2.3.8 GIS Requirements 23
2.3.9 Using Transit Competitiveness Index (TCI) 23
2.3.10 Using Sketch Planning Tool (SPT) 24
2.3.11 Traveler Attitudes 27
2.3.12 Demographics and Economics Inputs to the Structural Equation Model 27
2.4 Intelligent Transportation System Deployment Analysis System (IDAS) 29
2.4.1 Applications 29
2.4.2 Advantages 30
2.4.3 Disadvantages 31
2.4.4 Availability and Cost 31
2.4.5 Data Requirements 31
2.4.6 IDAS Analysis Process 32
2.4.7 Case Studies 33
2.5 Activity-Based Model 36
3.0 Transit Data Issues 40
3.1 Ridership Data 40
3.2 Running Time 40
3.3 On-time Performance 40
3.4 Sources of Errors for Ridership Data 41
3.5 Steps to Address Measurement Errors 41
3.6 Sources of Errors for Schedule Data 42
3.7 Relevant FTA Circulars 43
1.0 Introduction
With the building blocks of TRIPS123, SDP, and WDMS, and the commitment to develop and provide a state-of-the–art statewide multi-modal 511 service that appropriately showcases the unparalleled levels of transit service available in New York State, NYSDOT will be leading the development of a multi-agency Transit Service Information Portal (TSIP) to support a robust and flexible range of options for providing customers with comprehensive, high quality and useful information about their transit options and travel choices. It will also provide operators with an efficient and economical means of supplying standardized service data, in published open formats, to each other for coordination of services, to vendors deploying ITS applications that require schedule data such as AVL, to regional planning organizations, and to multiple downstream information outlets for marketing transit information.
This White Paper seeks to provide an Industry Scan of the new and emerging transportation planning tools being used in the industry and the transit data required for the tools.
1.1 Objective
The objective of the White Paper was to identify the state-of-the-art planning tools that are used for transit planning and discuss their strengths and weaknesses, transit data requirements, availability and cost.
1.2 Scope of White Paper
Specifically, the author interviewed several organizations that developed or adopted new tools to augment their current transportation planning activities. Through these discussions issues emerged on the transit data needs that drive the tools. The tools that are reviewed in this paper include:
· T-Best (Florida DOT)
· Puget Sound Sketch Planning Tools
· IDAS (US DOT)
· Activity-Based Model
1.3 Audience
The audience for this White Paper is the TSIP Real Time Regional Stakeholder Technical Working Group (RSTWG) members and interested public sector parties who will support the effort to move forward on identifying data needs for implementing application programming interfaces that support the upstream data loads and downstream uses of transit information to support regional planning. In order to fully understand the White Paper a basic knowledge of transportation planning and regional planning models, methods and concepts, and knowledge of transit data concepts are useful.
In addition, a short introduction to Regional Transit Planning is included in Section 2 to set the stage for a short discussion on how planning has evolved given the introduction and adoption of new planning tools. This introduction targets an audience who may not be quite as knowledgeable about regional transit planning.
2.0 Regional Transit Planning Tools
This section describes emerging trends and innovative regional planning tools used for transit planning activities.
2.1 Overview
Regional transportation and transit planning typically involves a wide range of activities including the development of travel demand forecasts, transit ridership forecasts, evaluation of current and future transit performance, economic and population forecasts, alternatives assessments for major transit and ITS/technology investments, market analysis, environmental analysis, consideration of equity issues and environmental justice, and fiscal analysis. These activities may be grouped into the following broad categories:
· Assessment of Current Service and Performance: The assessment involves monitoring existing conditions and service being delivered to transit customers. The elements evaluated can include transit accessibility, service (load, schedule adherence) and may include micro-level (route- or stop-level) analysis.
· Assessment of Alternatives for Major Investments: The impacts of changes in the transit network (route configuration, stop location), changes in transit service (frequency, availability), and changes in ITS/technology on accessibility, travel behavior, ridership, activity patterns, the environment and equity issues arising from the changes are evaluated. Project prioritization is facilitated by this assessment.
· Demographic, Economic, and Land Use Forecasting: Population and employment changes fuel increases or decreases in travel demand. Growth areas and corridors which may be conducive to transit are spurred on by land use changes;
· Development of Long Range Plans: Long range plans identify projected future transportation needs, capital improvement strategies to address the needs, and provides a vision of the future transportation landscape for the community. Long range plans are developed through information gathered from travel demand models and related models such as land use models, planning studies and market analysis, and stakeholder input. Long range plans may include plans for the introduction of new transit modes, deployment of ITS technologies, and other initiatives.
· Development of Short Range Programs: Short range programs are more focused on readily implementable operational strategies to meet regional objectives which usually include enhancing mobility, accessibility, safety, security, and efficiency.
· Environmental Analysis: Government through the U.S. EPA mandates the meeting of clean air targets. Environmental analysis involves assessing the impact of future improvements to a region’s transportation network on air quality and other aspects of the environment.
· Environmental Justice and Equity Issues: Environmental justice is concerned with the disproportionate negative effects of changes to the transportation network to the health of specific communities and neighborhoods. Questions such as – do the affected persons also benefit -- are also addressed. Equity issues answer questions such as - are low income and disadvantaged persons being impacted more by changes to the fare structure, schedules, transit network, etc.
· Development of a Fiscal Plan: A sound fiscal plan is needed to ensure that incoming revenues will meet all costs of proposed projects.
Figure 1: Metropolitan and Statewide Planning Flowchart
(FTA website: http://www.fta.dot.gov/planning/planning_environment_4160.html)
The flowchart of metropolitan and statewide planning (Figure 1) depicts the importance of determining and understanding a region’s vision and goals first, since all other steps are contingent about it. Then come the evaluation and prioritization of strategies and the development of a longer term transportation plan. The short-term TIP or Transportation Improvement Program(s) is created listing specific projects for the region in order of importance and finally the projects are completed and deployed.
Much of this regional planning which occurs within MPOs is focused around the travel demand model. Travel demand models are a cornerstone of long and medium range planning, and require vast amounts of data for them to generate useful information about future demand. The outputs of these models can then be used as input to post processor models and tools. Transit data required for the travel demand models are primarily transit networks, current ridership data, fare structure including transit fares and park and ride costs, and locations of park and ride lots. Other information and data typically utilized in travel demand models include demographics and socioeconomic indicators, and locations of any special demand generators.
More specifically, transit network includes route designations and patterns and may include the locations of transit stops or stations, park and ride lots, and transit amenities. Geographic coding or geocoding of these transit elements facilitates their use in GIS-based models and transit accessibility analysis. Assuming that the coding process is done conscientiously, it provides a common structure to an agency’s database of transit facilities and addresses problems regarding multiple names for the same stop or station. Ridership is typically defined as the number of unlinked passenger trips or passenger boardings, and is used to validate the demand model. Operations planners use passenger load information to determine necessary schedule changes or vehicle related changes. Fare structure affects demand and any discounted fares or free passenger types and parking fees at park and ride lots should also be incorporated. Schedule data: Schedule information is used in the mode choice element of demand models to estimate ridership and for monitoring and evaluating service delivered to customers which ultimately affects ridership levels. Additional transit service attributes may also be used in ridership forecasts.
2.1.1 The Travel Demand Model
The traditional four step travel demand model focuses on the aggregate trip-making characteristics of homogenous groups of travelers.[1] It has been used by MPOs for the last 50 years and consists of the following core steps:
1. Trip Generation – the first step predicts zonal trip productions and trip attractions; the region is typically divided into analysis zones and the land use characteristics of each zone then help determine the trips generated by each zone.
2. Trip Distribution – the second step estimates demand for specific destinations and matches origins with destinations; a trip table is created in this step. The attractiveness of a destination is determined by the gravity model which incorporates travel time from the origin and the quantity of attractions by destination.
3. Mode Choice – the third step calculates mode share; in this step, attributes of transit service are important. Attributes include schedules, availability, frequency of service, fare, accessibility, and capacity.
4. Trip Assignment – the fourth step assigns the trips to a specific route.
The demand model needs to be validated by comparing its results against current observed data. The following are data sources useful for calibration[2]:
· Accurate estimates of base year traffic analysis zone (TAZ) household characteristics and employment information
· An accurate representation of the base year highway and transit network
· An accurate base year travel survey, and
· Accurate base year ground counts are all needed for model calibration
MPOs have been adding supplemental models and data to enhance the prediction capability of their models. For example, Puget Sound has developed Land Use Allocation, Economic Forecasting, Vehicle Availability, and Truck models.
The four step model’s major shortcomings include the inability of the model to treat the impact of various factors such as ITS technologies on individual travel behavior. To address this issue, MPOs have been considering the development of activity-based models. Since the four-step model looks at aggregate numbers of trips and travel patterns, it sees linked segments of one trip as separate independent trips unlike the activity-based model which sees the segments as part of one continuous trip. There was also a realization that the model needs to be linked with simulation or dynamic traffic assignment models in order to evaluate policy.
The Activity-Based model models individual activity and travel behavior, and contains subsystems that forecasts population, activity and travel demand, interactions among vehicles and travelers, and air quality impacts.
Activity-Based models such as UrbanSim, TRANSMIS, and FAMOS have some similarities with the traditional four-step travel demand models. The trip generation step of the four-step model may loosely correspond to the activity generation phase of the activity-based model. Unlike the four-step model, the activity-based model monitors the travel patterns of individuals. Mode choice, like the four-step model, is user defined and appropriate impedance factors are applied. The route planning phase includes destination choice for each individual traveler and mode choice. The traffic or trip assignment step of the four-step model corresponds to the route planning and simulation elements of the activity-based model, and the path selection criterion is usually also shortest time. In addition, cost factors using the value of time are incorporated into activity-based models. There is also an emission module that determines the air quality impacts of the trips.
Activity-based models produced by different vendors may differ with regards to specific elements, especially since there are different ways to model activity and travel behavior. These methods include econometric or utility-based ways, rule-based processes, simulation, and psychometric methods. An example of a state-of-the-art activity-based model follows.
The Activity-Based Model TRANSIMS, for instance, is comprised of an Activity Generator, Route Planner, Traffic Microsimulator. The Activity Generator provides estimates of the number, characteristics, and locations of activities in which individuals will participate during the simulation period. The Activity Generator produces activities for each individual in a synthetic household based on surveys of actual household demographics and activity. Surveyed households with similar activity patterns are then grouped according to demographic attributes (age, gender, relationship). Individuals in the synthetic households are assigned an activity type, duration, mode preference, and start/end times based on information from the surveys. There is also a population synthesizer that is used as an input to the Activity Generator. Using the activity list provided by the Activity Generator, transit and network data, link travel times, and mode options for each link, the Route Planner selects the fastest routes for each tour of each individual in the synthetic household. The dynamic Traffic Microsimulator tries to capture the fluid nature of traffic behavior by simulating traveler interactions (e.g., congestion) and vehicle movements.