Simulated annealing algorithm for solving the capacitated vehicle routing problem: a case study of pharmaceutical distribution

Jurnal Sistem dan Manajemen Industri

Natasha UtamiSyaveyenda¹  FairuzKumari¹  Adji ChandraKurniawan¹  NandaRuswandi¹  Fiki RohmatulMaula¹  Anak Agung Ngurah PerwiraRedi¹  Annisa UswatunKhasanah² 

[1] Pertamina University;Universitas Islam Indonesia;

关键词: vrp;    nearest neighbour;    simulated annealing;    heuristic method;   

DOI  :  10.30656/jsmi.v4i1.2215

来源: DOAJ

【 摘 要 】

This study aims to find a set of vehicles routes with the minimum total transportation time for pharmaceutical distribution at PT. XYZ in West Jakarta. The problem is modeled as the capacitated vehicle routing problem (CVRP). The CVRP is known as an NP-Hard problem. Therefore, a simulated annealing (SA) heuristic is proposed. First, the proposed SA performance is compared with the performance of the algorithm form previous studies to solve CVRP. It is shown that the proposed SA is useful in solving CVRP benchmark instances. Then, the SA algorithm is compared to a commonly used heuristic known as the nearest neighborhood heuristics for the case study dataset. The results show that the simulated Annealing and the nearest neighbor algorithm is performing well based on the percentage differences between each algorithm with the optimal solution are 0.03% and 5.50%, respectively. Thus, the simulated annealing algorithm provides a better result compared to the nearest neighbour algorithm. Furthermore, the proposed simulated annealing algorithm can find the solution as same as the exact method quite consistently. This study has shown that the simulated annealing algorithm provides an excellent solution quality for the problem.

【 授权许可】

Unknown