| Abstrakt: |
Determining optimal routes is a very important part of the problem of optimizing transportation (distribution). Determining the optimal route can be modeled by applying graph theory regarding the study of the Vehicle Routing Problem (VRP). One variant of VRP is Multiple Trip Vehicle Routing Problem with Time Windows (MTVRPTW) which has the advantage that a vehicle can serve more than one route. In this article, the MTVRPTW problem is solved using the Adaptive Large Neighborhood Search – Tabu Search (ALNS – TS) algorithm with a Delphi 7 program design. The method used is waterfall, namely requirements, design, implementation, verification and maintenance. Results of the ALNS-TS algorithm Delphi program design on MTVRPTW with program input points indicating depot and customer, distance, demand, vehicle capacity, vehicle speed, service time, number of vehicles, parameters, TS iteration, ALNS – TS iteration, and time window. The program output is in the form of optimal route results with total travel time, distance, and graph visualization. The program design results were tested using black box testing, simulation data and standardized datasets. The results of testing the ALNS – TS algorithm on MTVRPTW with a standardized dataset have a small gap, namely 0.27%, which means that the program design is adequate to be applied to the problem of optimizing transportation problems. [ABSTRACT FROM AUTHOR] |
