Solving Multiple Depot Vehicle Routing Problem (MDVRP) using Genetic Algorithm

Abstract

Vehicle Routing Problem is a NP-Hard classical complex combinatorial problem described as task of determining efficient and shortest delivery or pickup routes to service several customers scattered in different geographical regions with a fleet of vehicles with additional predefined constraints to satisfy real-life scenarios. Vehicle Routing Problem has wide applications in Logistics and Transportation with growing economic importance. In Solutions of Vehicle Routing Problem maintaining the defined restrictions is of high Interest. Exact solutions of Vehicle Routing Problem can’t be obtained due requirement of high computation time. Due to Genetic Algorithm’s stochastic characteristics and efficiency in solving combinatorial problems it is used to find true and approximate solutions of Vehicle Routing Problem. For constrained variants of Vehicle Routing Problem feasible space is smaller than whole search space and Genetic Algorithm finds a solution with high precision that doesn’t violate any of the constraints.