Metaheuristics approaches to solve the employee bus routing problem with clustering-based bus stop selection

Vehicle routing problem (VRP) is a complex problem in the Operations Research topic. School bus routing (SBR) is one of the application areas of VRP. It is also possible to examine the employee bus routing problem in the direction of SBR problem. This chapter presents a case study for capacitated employee bus routing problem with data taken from a retail company in Turkey. A mathematical model was developed based on minimizing the total bus route distance. The number and location of bus stops were determined using k-means and fuzzy c-means clustering algorithms. LINGO optimization software was utilized to solve the mathematical model. Then, due to NP-Hard nature of the bus routing problem, simulated annealing (SA) and genetic algorithm (GA)-based approaches were proposed to solve the real-world problem. Finally, the performances of the proposed approaches were evaluated by comparing with classical heuristics such as saving algorithm and nearest neighbor algorithm. The numerical results showed that the proposed GA-based approach with k-means performed better than other approaches.