Perancangan Rute Distribusi Tabung Gas Industri Berdasarkan Konsep Pick Up and Delivery Menggunakan Metode Ant Colony Opyimization

Abstract

The imbalance between capacity and scheduled distribution routes causes delivery delays and inefficient use of time, which results in increased operational costs and decreased efficiency of distribution services. This study aims to design optimal distribution routes based on the Vehicle Routing Problem with Pick-Up and Delivery (VRPPD) concept by considering limited vehicle capacity and diverse delivery destinations. The method used begins with customer clustering using Nearest Neighbor, based on geographic proximity, customer demand, and vehicle capacity as an initial solution. Next, the route design is optimized using the Ant Colony Optimization (ACO) algorithm implemented with MATLAB software. The results show that the combination of clustering and the ACO algorithm can significantly reduce the total distance traveled and delivery time, directly impacting the efficiency of the distribution process. The route generated by the ACO method can reduce the total distance traveled by 21%. In addition, the total delivery time is also reduced by 16%. Thus, the application of the Nearest Neighbor clustering method and route optimization using ACO can significantly improve the efficiency of industrial gas cylinder distribution, both in terms of distance and delivery time. These findings provide a practical contribution to more effective and efficient distribution planning for companies facing vehicle capacity limitations and complex delivery needs.