Application of Statistical Process Control for Anomaly Detection in Automated Assembly Lines: A Case Study of Mikano International Limited

Manufacturing quality in automated assembly operations relies on the timely detection of process deviations before they produce large quantities of non-conforming output. This study applied Statistical Process Control (SPC) methods for anomaly detection on an automated assembly line, using Q-DAS software as the primary analytical tool. The work was carried out at Mikano International Limited, a power-generation equipment assembly facility in Lagos State, Nigeria. Five critical process parameters were selected for monitoring: generator frame dimensional tolerance, alternator winding resistance, terminal block torque, fuel line connection torque, and control panel switch actuation force. Measurement system analysis confirmed that all gauging systems had gauge reproducibility and repeatability values below 10%. X-bar and R charts were configured in Q-DAS procella for the three variable characteristics measured with subgroups of four. X-bar and S charts were applied to the two characteristics with subgroup sizes of eight. Control limits were calculated from 30 baseline subgroups collected under verified stable conditions. Western Electric detection rules and Nelson pattern tests were activated to extend chart sensitivity beyond three-sigma limit violations. Analysis of production data collected over a 12-week monitoring period showed that generator frame dimensional tolerance monitoring detected four process shifts, with an average detection lead time of 2.3 subgroups before confirmed non-conforming output was recorded. Alternator winding resistance monitoring detected three gradual drift events, two of which corresponded to confirmed field rewinding faults. The overall detection rate across all five characteristics was 88.6%, meaning that 31 of 35 confirmed process faults were preceded by a chart signal in Q-DAS. The false alarm rate across the monitoring period was 18.4%. Process capability analysis showed that three of the five characteristics achieved Cpk values above 1.33 after process adjustments, while two characteristics remained below 1.0. The study confirms that SPC implemented through Q-DAS is a practical approach for detecting anomalies in automated assembly operations.