The solution of location-routing problem with ant colony algorithm: Application in a food company

In this study, the capacitated location-routing problem was solved by using the real data of the company in the food industry. A new mixed integer programming model was developed by adding different constraints (warehouse capacity constraint and new objective function that includes only warehouse and route cost) to a mixed integer programming based location-routing problem in the literature. For large-scale problems, an ant colony of an algorithm, which is one of the meta-heuristic algorithms, was developed. To illustrate the effectiveness of the proposed algorithms, four scenarios were generated. By considering the MIP based model developed in the first scenario, the optimum result of the small size of the problem was found. In the second scenario, the large size of the problem is solved using the developed ant colony algorithm. Comparing the annual total cost of the company with the existing factory warehouse and the 5 proposed warehouses, it was observed that the logistics cost decreased by 49%. In the third scenario, the problem is solved by increasing the demands of certain provinces to show the effectiveness of the developed algorithm in determining new warehouse locations and new route. In the last scenario, the situation of opening branches in 10 new provinces within the next 5 years of the company was examined. In this study, it is seen that the algorithm works effectively and efficiently under the four different scenarios.