Optimisation of train scheduling in complex railways with imprecise and ambiguous input data by an improved integrated model

Modelling and scheduling of cargo and passenger trains with time limitations, queue priority and limited station lines is a cumbersome task. Furthermore, when distribution of traverse times are unknown or could not be estimated, conventional simulation approach fails and the need for a more intelligent approach becomes inevitable. This paper presents an integrated fuzzy modelling and simulation approach for such ambiguous cases. The case of this study is based on a specific and actual train route (800 kilometres). To show superiority of the approach the system under study was modelled through both conventional and fuzzy simulation approach. For the purpose of verification and validation, the fuzzy interval data of fuzzy simulation model was compared with the interval data of actual system through fuzzy analysis of variance (ANOVA). Then, random data from fuzzy simulation, conventional simulation models and actual system were compared with design of experiment (DOE). Hence, ANOVA and least significant difference (LSD) method were used to show fuzzy simulation approach is an ideal and superior modelling approach for uncertain and ambiguous scheduling railroad systems such as the actual case of this study.