| Abstract [eng] |
The research investigates and optimizes an automated screwdriving process utilizing the DEPRAG MINIMAT-EC-Servo 311ER27 screwdriving system. The system is implemented within an industrial production line used for assembling electronic components intended for the automotive industry. The need for this study arose due to an increased number of defective products, particularly during the production stage in which printed circuit boards (PCBs) are fastened to an aluminum base, which is positioned and secured inside a plastic housing. Since this stage directly affects product functionality, reliability, and manufacturing costs, its optimization is essential to ensure high quality and minimal additional losses. In the first part of the study, a comprehensive analysis of process data is carried out, examining over 300,000 screwdriving cycles, approximately 25,000 of which were identified as defective. Each cycle is characterized by key process parameters: torque angle, tightening torque, and overall torque. Based on this data, a defect classification is performed, and the main failure types are identified – overtightening, screwdriver bit slippage, and improper screw insertion angle. The root causes of these defects are analyzed through physical component assessments, including cross-sectional analysis, geometric measurements using coordinate measuring machines (CMM), and visual inspections. The study identified that the primary causes of defects are: inaccuracies in screwdriver positioning due to deviations in component alignment; mechanical damage to screws caused by excessive feeding speed or improper stopping mechanisms within the system; insufficient positioning capabilities of the screwdriver system. Based on the obtained results, four potential solution scenarios were proposed. A combination of these scenarios was used to develop the main solution. Main solution incorporates three system improvements: redesigning the screwdriving station by implementing two screwdrivers with independently controlled X-Y axes to enable precise tool positioning; introducing a mechanical calibration plate to ensure proper lift alignment; regulating airflow in the screw feeding system to reduce the risk of mechanical damage. These solutions are implemented in accordance with the project constraints – maximum spatial dimensions (1496 × 2530 × 2437 mm) and cycle time (≤10 s). By optimizing the screwdriving process and utilizing two screwdrivers, a cycle time of 4.9 seconds is achieved, meeting both technical and functional requirements. The redesigned components were validated using strength analysis based on the finite element method, which confirmed their reliability and suitability for operation without additional reinforcements. The analysis also revealed that the components could be further lightened without compromising system stability. This study highlights the importance of data-driven analysis, statistical process evaluation, and mechanical modeling in solution creation for automated systems. It emphasizes the advantages of an interdisciplinary approach – where mechanical design, control systems, and process parameter monitoring are integrated to ensure high-precision execution of automated operations. The presented methods and results provide a solid foundation for similar optimization projects in automated assembly systems. |
