Aiming at the planning of the number and location of daily necessities, this paper first proposes the principles for setting up the plan of material delivery. Select the regional isolated population, the area occupied by the region, the personnel mobility of material distribution, the total amount of food required by each person per day, and how many people can meet the daily needs of each living material delivery site as influencing factors, and collect factor data. The scientific planning algorithm of emergency materials based on fuzzy density peak clustering is established, and the original delivery quantity and location are optimized. The total number of optimized delivery points is 607, which only accounts for 39.01% of the original release sites. It is verified that the required number of delivery sites can meet the needs, the number of vegetable delivery points has been reduced by 949, but the average daily number of vegetables in each delivery site has not increased significantly. This algorithm can provide a scientific basis for managers on material planning problems.