<![CDATA[In the health sector, the availability of adequate medicines in pharmacies is very important in ensuring patients receive optimal care. The availability of drugs that is not well maintained can hamper the treatment process and have a negative impact on health services as a whole. The problem discussed in this research is the accumulation of drug stock caused by drug purchases that are not balanced with sales, causing losses to the pharmacy. Based on these problems, it can be stated that currently the pharmacy is not appropriate and effective in determining drug purchasing patterns. For this reason, it is necessary to determine drug purchasing patterns at pharmacies using the Apriori algorithm. This research aims to determine drug itemsets based on association rules so that these itemsets can be prioritized for stock in each purchase. This can also be displayed by an application prototype so that it is easier to get a combination of itemsets in determining drug purchases to help anticipate drug supply needs to be more efficient. . The final result of this research is a combination of itemsets in the form of drug items that meet the requirements for a minimum support value of 25% and a minimum confidence value of 60%, namely Methylprednisolone 4mg Novel and Paracetamol Mef with a support value of 41.57% and a confidence value of 62.50%. FG Troches and Paracetamol Mef with a support value of 25% and a confidence value of 100%, as well as Metformin 500MG Hj and Sanmol Tab with a support value of 25% and a confidence value of 60%. The final result of the association rules was an evaluation test to measure the strength of the relationship between items using the lift ratio and produced a value above 1%, namely an average test value of 2.4%, so it can be stated that the a priori results are said to be valid or strong.]]>
