Oil palm fruit theft at PT. Perkebunan Nusantara V still occurs frequently, especially in one district, Tandun District which has 6 oil palm and rubber plantations. The high rate of theft of oil palm fruit causes a decrease in the production of oil palm fruit and palm kernel. This study discusses the use of the K-Means Clustering algorithm as one of the data mining algorithms in grouping data on palm oil theft in the oil palm plantation area of Tandun District. With a number of theft data, this study enables the discovery of the potential for theft of oil palm fruit in the location of oil palm plantations at low, medium and high levels. K-Means Clusering algorithm calculation uses 3 supporting parameters, namely the area of each afdeling, the number of fruits that have been saved, and the number of fruits that have been stolen. After knowing the locations that have the potential for theft of oil palm fruit, the company can implement a strategy to safeguard those potential locations.