Transit authorities often prefer to outsource the bus routes to the private sector through a competitive tendering or negotiation procedure. A major challenge for decision-makers in this area is the optimal arrangement of routes among contracts which directly impacts the operating costs, labor costs, vehicle resources, level of competition between private operators, required subsidy, contract price, and system efficiency. Therefore, this study proposes a new model to find optimal route packages considering the parameters including package size, package revenue, the geographic distribution of routes, and terminal locations. The methodology relies on a genetic algorithm equipped with innovative methods, making it applicable to large-scale urban networks. We programmed both the model and the solution approach in C++ programming language and applied them to the city of Tehran as a real case study, which involved 169 routes and 2,215 bus fleets. In the main scenario, 10 packages were identified and compared with the current packages. The results demonstrated a 7.6% reduction in revenue deviation among the packages, indicating improved balance. Additionally, the routes within each package exhibited geographical concentration, aligning with the common terminal approach. By applying the proposed methodology to the bus network in Tehran, the study highlights the potential of the model in outsourcing bus routes to the private sector in similar real-world practices. It is worth noting that the Tehran Bus Organization has utilized the proposed packages derived from this study in their recent tender processes. Evaluation of the new packages' effectiveness demonstrated a notable decrease of 17% in average bid prices. This reduction directly corresponds to a decrease in operating costs and the required subsidy.
Read full abstract