INTERMODAL TRANSPORTATION NETWORK ANALYSIS – A GIS Application

Maria P. Boilé

Visiting Assistant Professor, Dept. of Industrial Engineering and

National Center for Transportation and Industrial Productivity

New Jersey Institute of Technology, Newark, NJ 07012.

ABSTRACT

A framework is presented which may be used to analyze and evaluate intermodal networks. An intermodal network can be defined as an integrated transportation system consisting of two or more modes. Modes on intermodal networks are connected through facilities which allow travelers and/or freight to transfer from one mode to another during a trip from an origin to a destination. The model may be used in the transportation planning field to make predictions regarding future network activity in terms of traffic volumes and travel costs, to evaluate alternative policies for mitigating congestion and reducing energy consumption and air pollution, and to aid the decision making process in terms of future transportation plans. To expedite and facilitate the effort of detailed intermodal network representation and analysis, the network model is interfaced with a Geographic Information System (GIS). This interface allows the user to take advantage of available sources of information on the physical components of the network, store the results of the proposed models in a GIS environment and display them in a spatial data format, and display several layers of information and create thematic maps to support planning and policy initiatives. The importance of the proposed interface as a delivery tool lies in the power of visual images to communicate technical information to non-technical people and simplify concepts. It may be employed to educate diverse audiences on issues such as how out travel habits affect the effectiveness of transportation systems, and the role of policy and decision-making in transportation.

Index terms—intermodal networks, transportation planning, Geographic Information System.

I. INTRODUCTION

The scope of the transportation planning process has changed from evaluating new regional transportation facilities to developing strategies that promote more efficient use of the existing transportation infrastructure while enhancing air quality. Most of the widely used applications were developed to analyze capacity expansions. These applications are not suitable for meeting the new requirements or performing tasks such as evaluating congestion pricing, transportation control measures, alternative development patterns or motor vehicle emissions (Shunk, 1992). The ability of existing theoretical developments to model intermodal systems and interactions between users and providers of transportation services is still very limited. Furthermore, the process of creating the input data files to be used along with these models is very time consuming and the outcome of the process is network specific. This paper presents a framework for analyzing and evaluating intermodal transportation networks using network equilibrium modeling in a Geographic Information System (GIS) environment.

Integration of intermodal network models with GIS provides a generic, user friendly functional environment to develop and analyze networks and evaluate the effects of policies and measures aimed to improve the service provided by one mode to the performance of other, competing modes. The importance of this research is reflected in some of the transportation-related provisions of the Clean Air Act Amendments (CAAA) of 1990, the Intermodal Surface Transportation Efficiency Act (ISTEA) of 1991 and its successor the Transportation Equity Act for the 21^st Century (TEA-21), which create a need for new and improved analytical tools. These tools will enable transportation planners to deal with intermodal issues and evaluate alternative policies to better utilize transportation systems. As an example, for passenger transportation systems, which are the focus of this paper, planners will be able to easily investigate programs aimed at reducing our reliance on single-occupant vehicles and making alternatives such as transit, high-occupancy vehicle lanes, bicycle and pedestrian facilities a more important part of the transportation program.

The next section provides some background information on intermodal networks, network equilibrium modeling and GIS. Section number three presents the modeling framework and discusses some of the issues in modeling intermodal networks in a GIS environment. Section number four presents an application of the proposed framework in modeling a transportation system. The last section provides a summary and concluding remarks.

II. Background

An intermodal network can be defined as an integrated transportation system consisting of two or more modes. In contrast to multimodal networks, modes on intermodal networks are connected through facilities which allow travelers and freight to transfer from one mode to another during a trip from an origin to a destination. Intermodal networks aim to provide efficient, seamless transport of people and goods from one place to another.

Transportation network equilibrium refers to the problem of users of a network seeking to determine their minimum cost paths between their origins and destinations (O-Ds). Network equilibrium models are used in the transportation planning field to make predictions regarding future network activity in terms of traffic volumes and travel costs, to evaluate alternative policies and to aid the decision making process in terms of future transportation plans.

A GIS can be defined as a computer system that can combine and analyze information from a database that has a spatial component. GIS technology relies on the integration of three aspects of computer technology: database management, spatial analysis tools, and graphics capabilities.

Numerous mathematical formulations and efficient algorithms have been developed to model network equilibria for passenger transportation networks. Evans, 1976; Florian and Nguyen, 1978; Smith, 1979; Boyce, 1980; Aashtiani and Magnanti, 1981; LeBlanc and Farhangian, 1981; Fisk and Nguyen, 1981; LeBlanc and Abdulaal, 1982; Dafermos, 1982; Boyce et al. 1982; Florian and Spiess, 1983; Boyce, 1984; Fisk, 1986; Boyce and Zhang, 1996 are some of the major contributions in this area. The models presented in these papers consider either one- or two-mode networks. When two or more modes are considered, traffic is assigned over modal networks. None of the models considers connections between modes and as a result they do not apply to intermodal networks.

The first network equilibrium models which explicitly consider and analyze intermodal networks are presented in Fernandez et al. (1994), and Boile et al. (1995). Fernandez et al. (1994) presents model formulations which consider two alternative modes available at each origin of the network. The alternatives are either auto and metro or auto and combined (auto-to-metro) modes. Combined modes are considered only at those origins where metro is not available. Boile et al. (1995) and Boile and Spasovic (1999) consider intermodal trips as an option at every origin of the network. This captures a common fact in most US urban areas, that even when a traveler has an option to walk to a nearby train station, he/she may prefer to drive to or be dropped off at another station along the route.

While these models have the ability to analyze intermodal networks and meet some of the legislative requirements, they are very data intensive as they require detailed data files on network characteristics to be created by the transportation planner. Generating these files is a very time consuming process. Furthermore, the outcome of this process is network specific, meaning that if the transportation planner wishes to analyze a different network, the same process has to be repeated. GIS technology provides a good solution to these problems, since it can take advantage of various sources of information on the physical components of a network which are available (TIGER/Line, 1995; US Streets, 1995; National Transportation Atlas Databases, 1997) for GIS applications. Use of GIS substantially reduces the efforts of data collection and modeling of network facility characteristics.

GIS applications with transportation planning capabilities such as TransCAD (Caliper, 1996) and UFOSNET (RST International Inc., 1998) are available today, however, these applications are limited in their ability to analyze intermodal networks since they typically assign traffic to distinct highway and transit networks without considering transfers between modes.

III. Modeling Framework

The framework presented in Figure 1 provides an efficient approach for modeling intermodal networks within a GIS environment. The method starts by generating within TransCAD, the GIS environment, the network specific data files necessary for the analysis. These files include information on network, service and demand characteristics, capacity of network facilities and cost components associated with the transportation services such as tolls, parking fees, transit fares. This information is then entered as input in a network equilibrium model such as the ones presented in Boile et al. (1995) and Boile and Spasovic (1999). The mathematical model statements developed within GAMS (Brook, Kendrick and Meeraus, 1992), a program used to formulate and solve complex mathematical programs, are generic. The network specific files are processed and classified in GAMS readable files and entered into the mathematical model through a Visual Basic (Microsoft Corporation, 1987-1998) program. Results of the network analysis provide information on traffic volumes and flow patterns on the network, as well as associated travel times and travel costs. These results are then exported back into TransCAD and used within cost functions to evaluate the impacts of traffic on the performance of the transportation systems. The input files can then be changed to reflect various operating and pricing policies, such as more frequent transit service, easier access to transit terminals, changes in tolls and parking fees, etc. Re-running the model with the modified input files will result in an evaluation of the alternative policies. The results, displayed in a spatial data format within GIS can be easily understood and interpreted. Several layers of information may be stored and thematic maps may be created to support planning and policy initiatives.

Figure 1. Modeling Framework

One of the challenges to the development of a GIS based intermodal network model was the combination of transit and road network layers into a single layer, as well as the introduction of intermodal transfer points, or stations. The procedure that was developed and used to generate connections between the various modes is shown in Figure 2.

Figure 2. Intermodal Network Connectors

The solid line represents a road network, and the rail line represents the transit network. The solid circle represents a trip origin or destination (a census zone centroid). The open circle represents an intermodal transfer point, or a train station. In Figure 2, "CD" represents a centroid link, connecting a point of origin or destination to the highway layer; "T" represents a transfer link, connecting the highway layer with transit stations; "FR" is a fare link, providing access to the station platform (by traversing this link, the traveler "pays" a fare for using the transit mode); and "W" is a walking link from an origin or destination to a transit station. This connection between the various layers gives a traveler the option to get off the highway, park at a train station and continue the trip using rail.

IV. Application

The framework presented in the previous section was used to model and analyze the Austin - San Antonio corridor in Texas. An overview of this network is shown in Figure 3. The user of the interface can visually select the network to be analyzed without having to manually enter information on the facilities of the network. The visual image and the database associated with it gives a better understanding of the network characteristics and minimizes the potential for error in generating the network data files.

Figure 3. Austin-San Antonio Corridor (overview)

The corridor was analyzed and the existing traffic conditions were modeled using the interface. Results of the analysis of the existing conditions (base case) are shown in Figure 4a. The links on the network are marked by the estimated traffic volumes moving between the two cities. These volumes represent the number of trips from Austin to San Antonio and from San Antonio to Austin. The main corridor is Interstate-35, which is heavily congested with through traffic (traffic that has its origin, destination, or both, outside of this network.

Figure 4. Base Case Model Results (graphical)

Without the GIS interface, the model results in terms of traffic volumes would be reported in the format shown in Table 1. The table show part of one of the model's output files. The first column lists the unique link ID number (according to the National Transportation Atlas Database; NTAD, 1997) while the second column reports the traffic volumes estimated by the network equilibrium model. The visual format of Figure 4 is very easy to understand and when several policies are analyzed, it makes it easier to compare the results.

Table 1. Base Case Model Results (table format)

Link ID Volume

543385 104.65

213396 0.00

213395 350.96

213394 76.99

213393 148.73

213392 18.21

213391 350.96

213390 350.96

… …

There are recent efforts in Texas to introduce passenger rail service along this corridor. The rail line is shown in Figure 4. This scenario, along with other alternatives may be analyzed and their results may be displayed graphically.

V. Conclusions

A framework for integrating intermodal network equilibrium models with GIS was presented. The integration provides a more efficient method for analyzing networks, by automating the inputs required for the models in a common environment and graphically displaying the model results. This integration may improve the ability of researchers and practitioners to quickly and accurately answer policy questions, evaluate competition between various modes and effectively communicate policy results to interested parties.

VI. Acknowledgment

This research was partially supported by the National Center for Transportation and Industrial Productivity, a member center of the University Transportation Centers Program and Lafayette College. This support is gratefully acknowledged, but implies no endorsement of the conclusions by these organizations.

VII. References

Shunk, G.A. (1992). Travel Model Improvement Program A Program Description, http://www.bts.gov /NTL/ page updated February 11, 1992.

Evans, S.P. (1976). “Derivation and Analysis of Some Models for Combining Trip Distribution and Assignment,” Transportation Research Vol 10, pp. 37-57.

Florian, M. and Nguyen, S. (1978). “A Combined Distribution Modal Split and Trip Assignment Model,” Transportation Research, Vol. 12-B, pp. 241-246.

Smith, M.J. (1979). “The existence, uniqueness and stability of traffic equilibria,” Transportation Research, 13B pp.293-304.

Boyce, D.E. (1980). “A Framework for Constructing Network Equilibrium Models of Urban Location,” Transportation Science, 14(1) pp.77-96.

Aashtiani, H.Z. and Magnanti T.L. (1981). “Equilibria on a Transportation Network,” Algebraic Discrete Methods 4 pp. 213-216.

LeBlanc, L.J. and Farhangian, K. (1981). “Efficient Algorithms for Solving Elastic Demand Traffic Assignment Problems and Mode Split-Assignment Problems,” Transportation Science, 15, No 4, pp. 306-317.

LeBlanc, L.J., and Abdulaal, M. (1982). “Combined Mode Split-Assignment and Distribution-Mode Split-Assignment Models with Multiple Groups of Travelers,” Transportation Science Vol. 16, No. 4, pp.430-442.

Fisk, C. and Nguyen, S. (1981). “Existence and Uniqueness Properties of an Asymmetric Two-Mode Equilibrium Model,” Transportation Science 15, pp. 318-328.

Dafermos, S.C. (1982). “The General Multimodal Network Equilibrium Problem with Elastic Demand,” Networks 12, pp. 57-72.

Boyce, D.E. (1984). “Urban Transportation Network Equilibrium and Design Models: Recent Achievements and Future Prospects,” Environment and Planning A, Volume 16, pp. 1445-1474.

Boyce, D.E. and Zhang, Y. (1996). “Combined Models of Nested O-D, Mode and Route Choice: Estimates and Comparisons for the Chicago Region,” Presented at the 75th Annual Meeting of the Transportation Research Board, Washington, D.C.

Florian, M. and Spiess, H. (1983). “On Binary Mode Choice/Assignment Models,” Transportation Science 17, No. 1, pp. 32-47.

Fisk, C. (1986). “A Conceptual Framework for Optimal Transportation Systems Planning with Integrated Supply and Demand Models,” Transportation Science, 20(1) pp.37-47.

Fernandez, E., DeCea, J., Florian, M. and Cabrera, E. (1994). “Network Equilibrium Models with Combined Modes,” Transportation Science 28, No. 3, pp. 182-192.

Boile, M.P., Spasovic, L.N. and Bladikas, A.K (1995). “Modeling Intermodal Auto-Rail Commuter Networks.,”Transportation Research Record, No. 1516, pp. 38-47.

TIGER/Line [cd-rom](1995). CD-TGR95-01. Issued: September 1996.

Available: U.S. Department of Commerce, Bureau of the Census, Geography Division, Washington, DC.

US Streets [CD-ROM] (1995). Caliper Data CD. Available: Caliper Corporation.

National Transportation Atlas Databases (1997) available through the BTS web site, www.bts.gov/gis/ntatlas.

Caliper (1996). TransCAD Transportation GIS Software User’s Guide, Version 4.0, Caliper Corporation.

RST International Inc. (1998) UfosNet Urban Travel Demand Modeling Software.

Boile, M.P. and Spasovic L.N. (1999). “An Intermediate Run Traffic Assignment Model for Intermodal Networks.” Submitted for consideration for publication in Transportation Research.

Brooke, A., D. Kendrick, A. Meeraus (1992) GAMS A User's Guide, Danvers, MA.