The development of an expert system for diagnosing pests and diseases of onion plants is of great importance given the significant role of these crops in the agricultural industry. This research aims to design and develop an expert system that can diagnose various pests and diseases that attack onion plants using the Naive Bayes method. This method was chosen for its ability to classify data based on probability assuming independence between features. This system is designed to assist farmers in identifying pests and diseases more accurately and quickly so that appropriate control measures can be taken immediately. The training data used in this study included symptoms that often occur in onion plants due to pest or disease attacks. Each symptom is associated with the probability of the appearance of a particular pest or disease. This expert system is designed with an easy-to-use interface for farmers, where they can enter the symptoms observed in plants. Based on these inputs, the system will analyze and provide a diagnosis along with recommendations for control actions that can be taken. The system testing results show that this expert system has good accuracy in diagnosing pests and diseases in onion plants. Thus, this system can be an effective tool for farmers in managing the health of their onion plants. Further research is recommended to improve disease and pest databases and expand the application of these systems to other plant types.