A Comparative Analysis of Performances of Econometric, Fuzzy and Time-series Models for the Forecast of Transport Demand

The paper attempts to evaluate how the fuzzy method can improve the range of forecast achieved by classic econometric and time-series models. Based on a survey of factors affecting rail passenger demand and using data of rail passenger demand of Greek Railways, the parameters affecting demand are identified with the use of the appropriate statistical testes. Using these parameters, econometric, fuzzy and time-series models are developed. A comparative analysis of the models developed and of their forecasting ability makes clear the range of applicability of each method and the accuracy and the reduction of ambiguity that can be reached with the use of the fuzzy method.