<![CDATA[The distribution of 3-kg LPG in the Greater Bandung area faces efficiency challenges in facility placement and service area allocation. This study aims to determine intermediate warehouse locations using the K-Means Clustering method and to allocate demand based on vehicle capacity. The clustering process was conducted using customer location coordinates, individual customer demand, and available vehicle capacity. This method was used to group 72 agents into balanced clusters based on geographical proximity and vehicle capacity constraints, ensuring that each cluster can be served by one vehicle without exceeding its maximum capacity. Based on a total demand of 5,570 units and a vehicle capacity of 560 units, this study produced 10 clusters with an average demand of 557 units per cluster. The centroid of each cluster is proposed as a candidate location for an intermediate warehouse. The demand reallocation process resulted in a more balanced distribution, as indicated by a reduction in inter-cluster demand variance from 41.9 to 3. These results indicate that the integration of K-Means Clustering and capacity-based demand allocation can be used as an initial approach in designing a more balanced and structured 3-kg LPG distribution network.]]>
Copyrights © 2026
