Penyelesaian Capacitated Closed Vehicle Routing Problem With Time Windows (Ccvrptw) Menggunakan Brkga Dengan Local Search

Determining the shortest route sequence for distributing productfrom depot to a number of outlets to minimize the total distribution cost is the main problem to deal with in Vehicle Routing Problem (VRP). Capacitated Closed Vehicle Routing Problem with Time Windows (CCVRPTW) is a variant of VRPthat accommodates the truck capacity and the working hours at the distribution. As CCVRPTW falls into NP-hard problem, it requires an efficient and effective algorithm to find the optimal solution. This research is aimed at designing Biased Random Key Genetic Algorithm (BRKGA) combined with a local search to solve CCVRPTW. The proposed algorithm is then coded in MATLAB. Using extensive numerical tests, the best setting of the algorithm parameters is obtained. The performance of the algorithm is then compared to a heuristic for solving a soft drink distribution problem. The result shows that BRKGA hybridized with a local search returns a lower total distribution cost compared to that of resulting from the heuristic. in addition, it is demonstrated that the proposed algorithm could further improve the performance of the standard BRKGA