Integrated Economic, Land Use and Network Growth Model for Transportation Management and Policy Analysis in the Washington DC Area

This paper demonstrates the feasibility of developing integrated urban systems model that considers transportation network investment and growth over time, which is a necessary tool for analysts to obtain accurate estimates of the total impact of transportation policies. Also, this paper presents a quantitative model that can forecast future networks under current and alternative transportation planning processes. The current transportation planning process is modeled based on empirical information collected from interviews with key transportation agencies and planning documents published by these agencies. The investment decision-making rules of and interaction/negotiations among state and local transportation authorities are explicitly considered in the proposed agent-based model. Results on a test network show the current transportation planning process can be improved in several different ways. Either a more centralized or more decentralized planning process can improve investment decision-making and enhance the performance of future transportation networks. While it is certainly feasible to employ the proposed model to evaluate alternative planning processes out of intellectual interests, the most likely practical application of this type of models is probably the evaluation of the impact of a particular group of investment projects on future network performance. Another application is to forecast future networks for long-range transportation planning and policy scenario analysis. Currently, there is not a general method for generating future transportation networks 30 or 50 years from now, though this kind of planning horizon is often required for land use, greenhouse gas, and sustainability policy analysis. The model developed in this paper can fill this methodological gap. Several aspects of the proposed model should and can be improved in future research. Model demonstration on a real-world network is clearly in order, and this work is underway for the statewide highway network in Maryland. The planning process model needs to be validated, possibly through comparisons between observed investment decisions and model estimated investment decisions. The current transportation planning process in other regions may also be studied and modeled.