A multi-commodity network flow model for railway capacity optimization in case of line blockage

In this study, a multi-commodity network flow model is proposed to optimize the railroad capacity under temporary line blockage. The proposed model enables the assessment of residual railroad capacity under heterogeneous traffic condition. The model searches for an optimized train timetable to maximize the number of possible train paths by maintaining acceptable percentages of delayed trains. Computational experiments are conducted on instances of Iran railway to evaluate the performance of the model regarding computational efficiency and solution quality. The outcomes demonstrate an average optimality gap of about 3.7% which quantifies the effectiveness of the optimization model within a reasonable computational time. The output of the optimization model has been compared with the UIC406 (International Union of Railway) standard. The proposed optimization model could generate a more realistic solution in comparison with UIC406 method. According to the obtained result, the maximum capacity of the rail line increases by approximately 26.7% compared with UIC 406 code.