Solving train formation problem using simulated annealing algorithm in a simplex framework

Abstract

SUMMARY The train formation plan (TFP) determines the train services and their frequencies and assigns the demands. The TFP models are often formulated as a capacitated service network design problem, and the optimal solution is normally difficult to find. In this paper, a hybrid algorithm of the Simplex method and simulated annealing is proposed for the TFP problem. The basic idea of the proposed algorithm is to use a simulated annealing algorithm to explore the solution space, where the revised Simplex method evaluates, selects, and implements the moves. In the proposed algorithm, the neighborhood structure is based on the pivoting rules of the Simplex method that provides an efficient method to reach the neighbors of the current solution. A state‐of‐the‐art method is applied for parameters tuning by using the design of experiments approach. To evaluate the proposed model and the solution method, 25 test problems have been simulated and solved. The results show the efficiency and the effectiveness of the proposed approach. The proposed approach is implemented to develop the TFP in the Iranian railway as a case study. It is possible to save significant time and cost through solving the TFP problem by using the proposed algorithm and developing the efficient TFP plan in the railway networks. Copyright © 2012 John Wiley & Sons, Ltd.

Multi-objective mathematical modeling of an integrated train makeup and routing problem in an Iranian railway company

“…Huntley et al [9] presented an MINLP model and used a SA algorithm to find a near-optimal solution. Adopting a hybrid approach, Yaghini et al [4] designed a SA which is based on simplex for the TMP, in which the moves in this method are evaluated, selected and implemented. Bagheri et al [10] proposed a method of placing hazardous material cars to minimize the derailment risk during the train assembly process.…”

“…Thus, one criteria to the maximization of the customers' satisfaction level and minimization of the costs is to assume maximizing trains in terms of the total frequencies, minimizing yard activities in terms of the total quantity of the action of shunting kind, and minimizing the train underutilization. This is currently missing from the literature (see for example [3][4]); basically, minimization of variable and fixed costs is common in the current literature. In addition, all the required constraints associated with train formation problem is not considered in the current stream of research.…”

“…In addition, such an optimization model also considers the operational and physical constraints of the network to achieve feasible and optimal sizes, frequencies, assignments, traction, and demand profiles regarding the specific objective functions [4]. However, due to NP-hardness issues, the main challenge remains to deal with the difficulty of solving the train makeup problem in the context of a design problem of a service network.…”

Simultaneous Train Rescheduling through Cancelling, Delaying and Re-Ordering Policies: Three-Phase Solution Method with Guaranteed Optimality

“…In a wide literature on rescheduling railway services, IP and MIP mathematical models have been used to formulate the problems [9,10], because they are able to accommodate the linearity of the objective functions and constraints. However, there are a lot of studies focused on proposing heuristic solution methods to solve the train rescheduling problems.…”

Automated Box–Jenkins forecasting tool with an application for passenger demand in urban rail systems

“…Mwakalonge and Badoe studied the estimation of the total travel demand of the region . The demand forecasting results can be an excellent source of information for the train formation plan designed to network service level and demand allocation as described by Yaghini et al . Integration of future demand information in the train formation plan can lead to further savings in time, cost, and emission while increasing customer service level.…”