P PT. Yasunli Abadi Utama Plastic is a manufacturing company in the field of injection molding that produces TRX3 helmets. PT. YAUP experienced problems with the high percentage of defects in the production of TRX-3 helmets, namely 5.785%. The purpose of this final project is to identify failures that occur during the production process using the Failure Mode and Effects Analysis (FMEA) method, to determine the appropriate solution/remedial solution to overcome the failures that occur and to design an expert system. The first step in this research is collecting data regarding the production process of the TRX-3 helmet and then forming an FMEA label to identify process failures and analyze the causes and consequences of each process failure that occurs during the TRX-3 production process. FMEA is based on the criteria of severity (S), occurrence (0), detectability (D) and the value of the Risk Priority Number (RPN) which is based on the multiplication between the values ​​of S,0 and D. After that, clustering is carried out to prioritize the values. The RPN should take appropriate recommended actions to overcome process failures that occur. The highest RPN value for a shell helmet is 448 with the type of process failure of the ABS raw material used whose composition is not in accordance with laboratory standards, the highest RPN value for the visor is 441 with the same type of process failure as the shell helmet, the highest RPN value for the bolt helmet is 240 with the type process failure of the oven engine temperature below 100°C, the highest RPN value for the peak helmet is 288 with the type of process failure the injection pressure is less than 50 Mpa & clamping force is less than 70 Mpa, the highest RPN value for the rear vent is 240 with the failure of the clamping process not locking tightly, the highest RPN value for the face shield is 270 with a failure to check the injection results which is less detailed, the highest RPN value for the TRX-3 helmet (assembly) is 224 with a failure of the semi-finished product inspection process that lacks detail. The results from the formation of the FMEA are then used to design an expert knowledge-based system or expert system with Winexsys software to facilitate the identification of defects that occur in the production process of the TRX-3 helmet. The identification process carried out for this expert system is based on the selection of the TRX-3 helmet component and then selecting the effect of the process failure that occurred then selecting the type of process failure and then selecting the cause of the process failure which in the end will come out a solution / solution that is in accordance with the process failure that occurred.