Rimbun Sarah Gloria, Sapti Wahyuningsih
Capacitated Vehicle Routing Problem with Time Window (CVRPTW) modeling is chosen if distribution routes with time window restrictions are used. The ALNS-TS algorithm is used to solve the CVRPTW problem by implementing the program using Borland Delphi 7. This ALNS-TS algorithm utilizes the Nearest Neighborhood stage as a Heuristic to build the initial circuit, followed by the Remove Request and Reinsertion Request stages. The final circuit is completed through Tabu Search calculation. The program design includes input of customer points, depot data, vehicle data, popsize parameters, and IterMax. The output includes total distance, travel time, and route graph visualization. The program was tested using simulated data of 7,25, 50, and 100 customers. Three parameters were tested, namely parameter p, tabu search iteration parameter (maxlt TS), and ALNS-TS iteration parameter (maxlt ALNS-TS). The results of the parameter test obtained that tabu search has a significant effect on the total route distance, where an increase in its value results in an increasingly optimal route. Meanwhile, the parameters p and ALNS-TS do not have a significant influence on the final route result. The program design results were also tested with the Solomon Dataset with codes C101_025 and C201_050, resulting in a better solution than the ALNS algorithm with a small gap. This indicates that the total route distance generated by the ALNS-TS algorithm is smaller than that of the ALNS algorithm. © 2025 American Institute of Physics Inc.. All rights reserved.
Department ofMathematics, Universitas Negeri Malang, Jl. Semarang 5, Malang, Indonesia
