A new Hybrid Genetic Variable Neighborhood search heuristic for the Vehicle Routing Problem with Multiple Time Windows

The vehicle routing problem is investigated by using some adaptations of the variable neighborhood search (VNS).The initial solution was obtained by Dijkstra’s algorithm based on cost network constructed by the sweep algorithm andthe 2-opt. Our VNS algorithm use several neighborhoods which were adapted for this problem. In addition, a number oflocal search methods together with a diversification procedure were used. The algorithm was then tested on the data setsfrom the literature and it produced competitive results if compared to the solutions published.

Download Full-text

Pengembangan Algoritma Hybrid Restart Simulated Annealing with Variable Neighborhood Search (HRSA-VNS) untuk penyelesaian kasus Vehicle Routing Problem with Time Windows (VRPTW)

Jurnal Rekayasa Sistem Industri ◽

10.26593/jrsi.v6i1.2427.49-56 ◽

2017 ◽

Vol 6 (1) ◽

pp. 49

Author(s):

Titi Iswari

Keyword(s):

Simulated Annealing ◽

Vehicle Routing ◽

Vehicle Routing Problem ◽

Variable Neighborhood Search ◽

Simulated Annealing Algorithm ◽

Time Windows ◽

Heuristic Method ◽

Neighborhood Search ◽

Routing Problem ◽

Time Required

Determining the vehicle routing is one of the important components in existing logistics systems. It is because the vehicle route problem has some effect on transportation costs and time required in the logistics system. In determining the vehicle routes, there are some restrictions faced, such as the maximum capacity of the vehicle and a time limit in which depot or customer has a limited or spesific opening hours (time windows). This problem referred to Vehicle Routing Problem with Time Windows (VRPTW). To solve the VRPTW, this study developed a meta-heuristic method called Hybrid Restart Simulated Annealing with Variable Neighborhood Search (HRSA-VNS). HRSA-VNS algorithm is a modification of Simulated Annealing algorithm by adding a restart strategy and using the VNS algorithm scheme in the stage of finding neighborhood solutions (neighborhood search phase). Testing the performance of HRSA-VNS algorithm is done by comparing the results of the algorithm to the Best Known Solution (BKS) and the usual SA algorithm without modification. From the results obtained, it is known that the algorithm perform well enough in resolving the VRPTW case with the average differences are -2.0% with BKS from Solomon website, 1.83% with BKS from Alvarenga, and -2.2% with usual SA algorithm without any modifications.Keywords : vehicle routing problem, time windows, simulated annealing, VNS, restart

Download Full-text

PENERAPAN SIMULASI DAN RELIABILITAS PADA MODEL VEHICLE ROUTING PROBLEM (VRP) DENGAN PERMINTAAN PROBABILISTIK

Simetris Jurnal Teknik Mesin Elektro dan Ilmu Komputer ◽

10.24176/simet.v9i1.1583 ◽

2018 ◽

Vol 9 (1) ◽

pp. 189-204

Author(s):

Ary Arvianto ◽

Rizal Luthfi Nartadhi ◽

Diana Puspita Sari ◽

Wiwik Budiawan

Keyword(s):

Vehicle Routing ◽

Vehicle Routing Problem ◽

Time Windows ◽

Multiple Time ◽

Routing Problem ◽

Fleet Size ◽

Split Delivery ◽

Multiple Product ◽

Multiple Time Windows ◽

Fleet Size And Mix

Vehicle Routing Problem (VRP) memiliki aplikasi yang penting di bidang manajemen distribusi, sehingga menjadi salah satu contoh masalah yang banyak dipelajari dalam literatur optimasi kombinatorial dan diakui sebagai salah satu pengalaman tersukses dalam riset operasi. Dalam penelitian ini dilakukan simulasi dari penelitian [1] dengan memperhatikan varian VRP homogeneous fleet size and mix vehicle routing, multiple trips, multiple product and compartements, split delivery, dan multiple time windows dan permintaan tidak pasti (Probabilistic Demand). Hasil yang didapatkan bahwa model yang dibuat telah mampu merepresentasikan penelitian sebelumnya [1] dengan verifikasi hasil yang sama. Permintaan tidak pasti ditunjukkan dengan melakukan pengurangan kapasitas sebesar 5% dan 10% dengan hasil bahwa dengan mengurangi kapasitas sebesar 5% terjadi permintaan pelanggan yang tidak tercukupi di beberapa pos, sedangkan pada 10% semua permintaan dapat tercukupi. Namun dari segi biaya pada nilai 10% memiliki biaya yang lebih tinggi daripada 5% hal ini dikarenakan rute yang dihasilkan lebih banyak sehingga mengakibatkan penggunaan kapal lebih banyak dilakukan.

Download Full-text