Development of heuristic methods based on genetic algorithm (GA) for solving vehicle routing problem

Abstract

The Vehicle Routing Problem (VRP) is an important area and has been studied as combinatorial optimization problems. VRP calls for the determination of the optimal set of routes to be performed by a fleet of vehicle to serve a given set of customers. VRP in which demand at each location is unknown at the time when the route is designed but is follow a known probability distribution, is known as VRP with Stochastic Demands (VRPSD). VRPSD finds its application on wide-range of distribution and logisticstransportation sector with the objective is to serve a set of customers at minimum total expected cost. One of the applications of VRPSD is in the case of picking up garbage done by solid waste collection company. The computational complexity of most vehicle routing problem and moreover the intricate of stochastic VRP algorithm has made them an important candidate for solution using metaheuristics. This research proposes the enhanced metaheuristic algorithms that exploit the power of Tabu Search, Genetic Algorithm, and Simulated Annealing for solving VRPSD. Genetic Algorithm as population-based methods are better identifying promising areas in the search space, while Tabu Search and Simulated Annealing as trajectory methods are better in exploring promising areas in search space. Simulated Annealing is a global optimization technique which traverses the search space by generating neighboring solutions of the current solution. A superior neighbor is always accepted and an inferior neighbor is accepted with some probability. Tabu Search is similar to Simulated Annealing, in that both traverse the solution space by testing mutations of an individual solution. However, simulated annealing generates only one mutated solution but Tabu Search generates many mutated solutions and moves to the solution with the lowest fitness of those generated. Genetic Algorithm gives a pool of solutions rather than just one. The process of finding superior solutions mimics the evolution process, with solutions being combined or mutated to find out the pool of solutions. This research explored and developed new heuristics based on GA for solving VRPSD. New algorithms, journal papers and computerized system were also developed. Future, area that may be explored include the used of Ant Colony Optimization (ACO) which exploits the nature phenomenon of ants. Based on the proposed heuristic method, we developed a program to optimize the routing problem using the Visual Studio C++ 6.0 programming language.