Optimasi Pelayanan Kapal Penumpang melalui Clustering Penumpang dengan Metode Silhouette Coefficient

Abstract

Improving the efficiency of passenger ship services is a big challenge for the shipping industry. This research aims to optimize ship passenger services at Muara angke port by segmenting passengers using the Silhouette Coefficient method. This research uses the k-means algorithm and the Silhouette Coefficient method in order to get passenger clusters based on boarding passengers and disembarking passengers. The Silhouette Coefficient method in this research is used to find the number of clusters. The dataset used is passenger data from January 1 - December 31, 2021. The results showed that 2 clusters were obtained, namely the highest cluster (C1) and the lowest cluster (C2) from the k-means algorithm with the results of the centroid value obtained for cluster 1 passengers up, namely 53,414 and passengers down, namely 54,585, the centroid value obtained for cluster 2 passengers up, namely 596,396 and passengers down, namely 532,455. The value with the Silhouette Coefficient method is obtained 0.704 in cluster 1 for the highest cluster there are 2091 data with a percentage value of 57.27% and 0.548 in cluster 2 for the lowest cluster there are 1560 data with a presentation value of 42.73%. The Silhouette Coefficient value is close to the value of 1, which means that the Muara angke port is good enough in optimizing passenger services.