Drug supplies are an important aspect because of their large value and large quantity and are an important factor in supporting health services in community health centers. Ineffective drug management, especially in terms of needs planning, can lead to excess or shortage of stock. Both conditions have negative impacts, such as budget waste, drug expiration, or even disruption of patient services due to unavailability of drugs. At the Pante Bidari Health Center UPTD, the drug needs planning process is still carried out manually or based on rough estimates without using sophisticated technology. This study aims to design and build a web-based drug inventory prediction system using the Least Square method. The Least Square method was chosen because it is able to carry out the forecasting process quickly and with good results. In this study, the type of data obtained is drug usage data, data is grouped based on each supplier, from the health center information system during a certain period. After going through the pre-processing and calculation stages, the predicted values are calculated and displayed through a web-based system designed to be easy to use by health center officers. The web system developed in this study uses PHP as the programming language and MySQL as the database, implementing the Least Square method effectively. The results of this study are a drug usage prediction application for the future, applying the Least Square method, which displays drug usage data over a certain period. The system will present the data in the form of a table. Based on testing the drug usage data for Acyclovir Cream 5 mg from January 2023 to August 2024, the prediction result for the following month, September 2024, is estimated to be 38.415, which is rounded to 38 units of the drug.