Modelling of route choice behaviour with fuzzy logic approach

Traffic patterns which are defined according to the decisions about route choice and mode choice are very important part of daily life. The route choice is a complex problem. It depends on many parameters due to individual’s utility maximization. The effectiveness of a route choice model is characterized by individual’s perception which changes in wide range; hence, the phenomenon has to be well understood. The right route choice decision has crucial importance both for the road users to get the highest benefit and for traffic planners to form the effective traffic assignment. The purpose of this study is to model the route choice problem with Fuzzy Logic technique (FL), by using the real data obtained by a survey. The real time data is calibrated with artificial neural networks to create optimum fuzzy membership functions. A fuzzy logic model that uses the five most important parameters of route choice behavior; travel time, traffic safety, congestion, cost and environmental factors are proposed. Based on a set of fuzzy ‘if–then’ rules is developed to represent a typical driver's psychology for capturing essential preferences, pairwise, among the alternatives that a driver may consider. It is aimed to develop a new approach to solve the problems of imprecision, vagueness and uncertainty characteristics of route choice problem for making possible a more proper description of route choice behavior in transportation systems. Findings show that this method provides intuitively promising results.