T The department of cooperatives, smes, trade, and industry of merangin regencypossesses comprehensive small and medium industry (smi) data; however, it iscurrently presented in static table format, which is insufficient to effectively supportanalytical processes and decision-making. this study aims to develop an interactivedashboard based on microsoft power bi that presents smi information visuallywhile simultaneously grouping sub-districts according to their smi characteristicsusing the k-means algorithm.the data used in this study consists of 2,240 smi business units from meranginregency in 2024. the research process includes data aggregation at the sub-districtlevel, normalization using min-max scaling, determination of the optimal numberof clusters using the elbow method, clustering with k-means, and evaluation usingthe silhouette score. the clustering variables comprise the number of smis, totalworkforce, number of industry types, and average workforce per smi.the data was successfully aggregated into 24 sub-districts, with the optimal numberof clusters determined to be three. a silhouette score of 0.434 indicates anacceptable clustering quality. the resulting dashboard is capable of presenting smidistribution, cluster characteristics, and industry type distribution interactively,thereby serving as a supporting tool for local government in formulating moretargeted smi development policies.