MATHEMATICAL MODEL AND HYBRID DISCRETE PARTICLE SWARM OPTIMIZATION ALGORITHM FOR VEHICLE ROUTING PROBLEM WITH HETEROGENEOUS VEHICLES, MULTIPLE PRODUCTS, MULTIPLE COMPARTMENTS, MULTIPLE TRIPS, SPLIT DELIVERY, AND MULTIPLE TIME WINDOWS

• Main Author: Yulianto Putra, Dani
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/84646
• Institution: Institut Teknologi Bandung

This study discusses vehicle routing problems with heterogeneous vehicle, multi- product, multi-compartment, multiple trip, split delivery, and multiple time windows (VRP-HMPMCMTSDMTW). This research develops a mathematical model and algorithm for the VRP case that was carried out by Marcellina (2019) by developing vehicle characteristics to be heterogeneous. The model developed in this research is to solve the real case problem of fuel distribution by PT. Pertamina in East Nusa Tenggara and Timor Leste areas. The mathematical model developed in this study is in the form of Mixed Integer Linear Programming (MILP) which has performance criteria of minimizing total vehicle operational costs in fulfull all customer demand during the planning horizon. The model solution was completed with the help of LINGO 18. In addition, a hybrid discrete particle swarm optimization (hDPSO) algorithm was also developed to overcome the problem of computing time in analytical methods. The results obtained are that the mathematical model developed can solve the VRP- HMPMCMTSDMTW and the hDPSO algorithm developed can produce a feasible solution for 5-8 customer data, with a gap of 1,85%. The developed model can also be used in vehicle routing problem with homogeneous vehicle, single product and single compartment, and single time window.