The factors affecting early decommissioning of distribution network equipment are complex and diverse, and multiple factors interact with each other. In order to filter out the main candidate set of factors affecting early decommissioning of equipment, data mining algorithms can be used to obtain association rules among them. Among them, Apriori algorithm is the most classical algorithm for mining association rules. However, due to the time complexity of the traditional Apriori algorithm is too large, the computational efficiency is not high. To address this situation, an Apriori optimization algorithm based on three-dimensional matrix is proposed to reduce the computational redundancy in the traditional algorithm by establishing a three-dimensional matrix as well as a parsimonious database to mine the frequent item set of factors affecting the early decommissioning of distribution network equipment. By comparing the experimental results, it can be seen that the improved algorithm can obtain the association rules of the decommissioning factors of the distribution network equipment and significantly improve the computational efficiency.