Hybrid Simulated Annealing Variable Neighbourhood Search Untuk Permasalahan Vehicle Routing Problem With Time Windows (Studi Kasus : PT. Joyful Trans)

Prasono, Danu Yudhi (2016) Hybrid Simulated Annealing Variable Neighbourhood Search Untuk Permasalahan Vehicle Routing Problem With Time Windows (Studi Kasus : PT. Joyful Trans). Masters thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
2513203203-Master-Theses.pdf

Download (2MB) | Preview

Abstract

Permasalahan transportasi saat ini semakin meningkat dan memberikan dampak yang cukup besar pada konsumsi sumber daya untuk dijadikan sebagai fasilitas utama dan pendukung. Moda transportasi darat berupa perusahaan travel di Indonesia memiliki permasalahan yang cukup kompleks dan termasuk dalam permasalahan NPhard. Sejauh yang peneliti ketahui belum ada penelitian mengenai hybridizing Simulated Annealing (SA) dengan Variable Neighbourhood Search (VNS) untuk permasalahan Vehicle Routing Problem with Time Windows (VRPTW). Tesis ini menggabungkan Simulated Annealing (SA) dengan Variable Neighbourhood Search (VNS) menjadi algoritma SAVNS dengan harapan menghasilkan solusi yang cepat dan stabil (steady state). Penggunaan mekanisme local search pada VNS ke dalam algoritma SA menghasilkan algoritma yang baru dalam bidang keilmuwan untuk menyelesaikan permasalahan VRPTW. Hasil algoritma diuji pada studi kasus pada perusahaan travel di PT. JOYFUL TRANS. Ukuran kinerja yang digunakan pada tesis ini adalah biaya dan rute optimal dengan jarak terpendek yang dihasilkan. Solusi akhir dari penelitian ini memberikan solusi rata-rata waktu komputasi sekitas 1,11 detik dan tingkat penghematan biaya sekitar 11,36% untuk permasalahan 8 titik. Sedangkan, permasalahan 14 titik memberikan hasil dengan rata-rata waktu komputasi sekitar 84,8 detik dan tingkat penghematan biaya sekitar 22,42%. ==================================================================================== Nowadays, transportation problem is going to rise up and give enough big effect to consumption of resource for making it as a major and support facility. Land transportation modes such as travel and transportation agents in Indonesia have a quite complex problem and that belongs to NP-hard problem. As far as researcher’s knowledge, there are no such research about hybridizing Simulated Annealing (SA) with Variable Neighbourhood Search (VNS) for Vehicle Routing Problem with Time Windows (VRPTW). This dissertation combine Simulated Annealing (SA) with Variable Neighbourhood Search (VNS) to a new algorithm that is called SAVNS, in order to obtain fast and steady state solution. The use of VNS local search mechanism in SA algorithm produce a new approachment in scientific to solve VRPTW case study. The output of the algorithm is going to be tested in case study of PT. JOYFUL TRANS travel agent. The use of performance measurement in this dissertation are minimum cost and shortest route with optimal solution. The final result for this research make solution with average computational time around 1,11 seconds and cost reduction rate around 11,36% for eight nodes problem. However, fourteen nodes problem give a result with average computational time around 84,8 seconds and cost reduction rate around 22,42%.

Item Type: Thesis (Masters)
Additional Information: RTI 511.8 Pra h
Uncontrolled Keywords: Simulated Annealing; Variable Neighbourhood Search; Vehicle Routing Problem with Time Windows
Subjects: T Technology > TE Highway engineering. Roads and pavements > TE7 Transportation--Planning
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions: Faculty of Industrial Technology > Industrial Engineering > (S2) Master Theses
Depositing User: Mrs Anis Wulandari
Date Deposited: 13 Jun 2017 04:19
Last Modified: 27 Dec 2018 04:19
URI: http://repository.its.ac.id/id/eprint/41627

Actions (login required)

View Item View Item