An intelligent auction-based capacity allocation algorithm in shared railways

Despite the theoretical foundations and successful practical applications of auctions in infrastructure allocation, such as radio waves, wireless communications, airports, and ports, the use of this method in allocating rail infrastructure capacity is limited. This paper addresses this gap by proposing an intelligent auction-based capacity allocation algorithm for railways. Unlike previous timetable-based models, the algorithm utilizes a two-objective integer optimization model, offering a practical solution for real-world problems. An additional contribution of the proposed model is the consideration of two crucial objectives: increasing revenue (or profit) and improving freight transportation performance, which have been overlooked in previous algorithms. The algorithm incorporates the enhanced constraint method to generate the optimal Pareto curve. Empirical results from the Iranian railway experimental network demonstrate the algorithm's efficiency, with performance improvements of 6% or more and over a 150% increase in the profit margin of the infrastructure manager compared to the current traditional method.