OPTIMIZATION OF PRODUCT STORAGE PLACEMENT IN WAREHOUSE USING PARTICLE SWARM OPTIMIZATION ALGORITHM, EXTENDED WEIGHTED TREE SIMILARITY ALGORITHM, AND GENETIC ALGORITHM METHODS

• Main Author: Zidni, Hasan
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/78865
• Institution: Institut Teknologi Bandung

E-commerce business has experienced significant growth in accordance with high sales transactions. This has an impact on reducing the public's need for retail stores, yet the need for modern warehousing and distribution centres is increasing. Therefore, a smart warehouse system is required to increase the effectiveness and efficiency of warehouse activities. One component of warehousing activities that cannot be removed from warehousing activities and has high costs and time is goods movement activities (receipt to storage location and storage location to delivery). Therefore, based on the problem, we propose determining storage location based on performance functions (warehouse policy and warehouse conditions: storage distance, delivery distance, total movement distance and storage line availability). Consequently, this research proposes an algorithmic method for making decisions on storage location, namely the extended weighted tree similarity algorithm, the PSO algorithm, and the genetic algorithm. In the third test, the proposed algorithm is based on 2 warehouse conditions (before and after the products are placed in the warehouse). The results of the test show that the extended weighted tree similarity algorithm and the PSO algorithm provide a minimum total movement distance. Meanwhile, the genetic algorithm results in a total moving distance that is not much different from the two algorithms. Furthermore, the PSO algorithm shows the fastest average computing time of the 2 warehouse conditions, approximately 7.70 seconds, followed by the genetic algorithm with an average of 15.84 seconds and the extended weighted tree similarity algorithm with an average of 24.22 seconds. Besides testing several algorithms separately, combining the PSO algorithm with the ABC method helps in grouping products with high to lowest demand. Although the results from the combination of the PSO algorithm with the ABC method are not as good as the PSO algorithm itself, the combination of these algorithms helps in product grouping.