Human Cognitive Reliability and R&D Efficiency: A Human Factor Study in Semiconductor Pilot Scale Production Line

A qualified chip sensor determined for mass production must undergo rigorous development and verification on the pilot line. In this experimental process, the sensor performance, quality, and R&D efficiency deeply impact the subsequent progress in the industry. Most pilot lines rely on human(engineers)-machines collaboration because the full-automatic production line requires extremely high investment and does not meet the demand for flexible technique adjustments. Therefore, the operational differences of individual engineers will directly affect the product’s qualification rate. It requires the Process Design/Integration Engineers (PIE, as a manager role of the fab) to update the process flow or management strategies to improve the operator’s performance and reduce human error probability, thus lowering the time and economic cost. For exploratory purposes, this study investigated three Semiconductor pilot scale production lines and interviewed 17 participants with multiple roles in the fab. It summarized 10 Common Performance Conditions and the related secondary index for operators’ performance evaluation and discussed its results in relation to potential human reliability errors.