An adapted multi-objective genetic algorithm for solving the cash in transit vehicle routing problem with vulnerability estimation for risk quantification

This study aimed to develop a model for vehicle routing problem with two objective functions of risk and distance minimization to optimize safety of cash/valuable commodities transportation. It is necessary to properly anticipate and prevent the occurrence of robbery to reduce the vulnerability to robbery attempts. The proposed approach for the vulnerability estimation of an armed robbery has been based on game theory and multi-criteria decision making (MCDM), which can accurately measure the amount of risk. A new multi-objective intelligent genetic algorithm (MOIGA) comprised of various heuristics is also designed to identify and intelligently select the most efficacious heuristic. The following experiments are used to test the proposed MOIGA: (1) Examining the influence of each proposed operator on the performance of the algorithm; (2) Evaluating the quality and diversity of MOIGA solutions compared to other popular algorithms. The obtained results demonstrate the effectiveness and efficiency of the proposed algorithm.