Rail versus Pipeline: Bayesian Decision Problem

Abstract

In response to increased foreign demand for United States bulk commodities, state officials must decide how to improve their port facilities and inland transportation networks. They must consider where improvements should take place, the size of the improvements, and how modes such as railroads and pipelines should be used. They are uncertain, however, about many things: the stability of export markets and their state's share; the cost of the transportation projects; the future payoff to the state in terms of revenues, jobs, and industrial tax base; and the negative impacts on the environment, recreational assets, and quality of life in the port area. An approach to solving this problem based on the theory of Bayesian decision analysis is presented in this paper. The approach is sufficiently general so that it may be applied in a variety of settings where port development is being contemplated, and it is demonstrated with a specific case study based on recent North Carolina coal export projects. Results of the analysis indicate that a 1981 decision to carry low volumes of coal by rail to the state port in Morehead City was correct and that, if the demand for North Carolina coal exports exceeds 3,000,000 annual tons, then a pneumatic pipeline should be considered.