<![CDATA[A good distribution is one of the company's strategy to increase the productivity of the company. Distribution strategy is indispensable in bottled drinking water, because bottled water business has increased every year. Distributor of bottled water has a variety of types and brands of goods with different packaging forms. The number of shipping destinations poses many problems in the distribution process, because it takes more time to arrive at different address and distances. This research has a goal to create a system that can help the process of distribution of goods with number of sales more than one, the problem is called Multiple Traveling Salesman Problem (M-TSP). One method to solve M-TSP problem is to use genetic algorithm, so it can determine the route with the shortest distance that will be visited by every sales. The genetic algorithm process uses permutation representations with chromosome length according to many customer orders and the number of orders on each sales, each gene is a number representing the customer number and the number of orders that each salesperson should visit. The test results show that the route sequence generated from the application of the genetic algorithm is better than that applied to the distributor with a total distance of 89.3 km and the fitness difference is 10.656578. The optimal parameters were obtained by generating population size 180, 400 generation and crossover rate 0.6 and mutation rate 0.4.]]>
Copyrights © 2017
