Metal contamination in plastic resin is a critical issue in automotive injection moulding production systems. This contamination leads to product defects, check ring damage, cushion value instability, and high maintenance costs. This study aims to implement a continuous improvement approach using the A3 Problem-Solving method, centred on the design and implementation of a Programmable Logic Controller (PLC) based automated Preventive Magnetic Cleaning (PMC) system. This study addresses findings regarding design gaps in a PLC-based (Mitsubishi) self-cleaning magnetic filter system, integrated with the A3 problem-solving framework. The system automatically executes a cleaning cycle whenever material transfer occurs, thereby proactively preventing the magnetic shielding phenomenon without requiring manual intervention. The A3 framework was utilized to systematically identify the root cause via fishbone analysis, revealing the magnetic shielding phenomenon as the primary failure mechanism. As a technical solution, a neodymium magnet-based self-cleaning filter system was designed and installed on the auto-feed line of an 850T servo injection machine, controlled by a Mitsubishi PLC using ladder logic. Results demonstrated the elimination of manual cleaning interventions (2 hours/cycle), stabilization of cushion values from an average of 4.2 to 8.4 (within the standard range of 6–8), the capture of 100 grams of metal contaminants per week, and savings of IDR 47,000,000 in check ring replacement costs, with a payback period of 0.4–1.5 months. This study highlights the effectiveness of integrating lean management tools with automation technology as a sustainable continuous improvement strategy.