Analysis of investment in private toll road

Participation in toll road investment by private firms becomes very popular nowadays. As there are more and more toll roads operating together on the same network, the chance to have impacts with the other operators also increases. Especially when the toll roads are serving for the same demand, their profits are interrelated due to demand inter-dependence in the network. A competitive game model is developed in the first part of this thesis to analyze the strategic interactions between the private toll road operators in determining their road capacity and toll level over the network. A simple but representative case of two competitive firms, each providing a single toll road on the network, is considered where the two toll roads are either substitutable or complementary in terms of their demand interdependence. A Quasi-Newton method in conjunction with sensitivity analysis method of equilibrium network flow is used to determine the competitive game solutions subject to network equilibrium constraints. On the other hand, investment in road project always involves a variety of uncertainties in the future, such as population growth rate, and other unexpected factors. It is important for the investors to have a clear picture about the impacts of these uncertainties on their expected returns before they can make any decision. In the second part of this thesis, we will focus on the effects by the uncertainty in potential demand. The growth of potential demand is assumed to undergo multiplicative Geometric Brownian shocks, a stochastic process. Monte Carlo technique is applied to forecast the future variables. Graphical representation of the result is provided for intuitive understanding of the effects on the variables, such as their deviation and confidence intervals. The impacts on the annual revenue, accumulative profit and breakeven year are discussed. The simulation results can provide the investor a reasonable investment strategy with acceptable confidence interval.