An Optimized Model of Bus Scheduling Based on Actual Variable Costs

Bus systems play an important role in urban public transport, and bus schedules are essential to bus companies and passengers. Historically bus timetables have generally been set manually, which is inefficient. In this paper a new model aiming to help bus companies with bus schedule will be developed. In short, the number of passengers at each station within a given time is firstly counted through IC card data from the bus IC card charge system; then by taking the actual variable costs including bus companies’ operation costs and passengers’ waiting cost into consideration, a new model is proposed to calculate optimal bus departure intervals. A genetic algorithm is adopted to get the optimal schedule quickly and precisely. In addition, it is found that the efficiency of urban bus schedule in the test scenario can be further improved compared to the original schedule, which supports the need for future research.