An approach for grouping circuit cards into families to minimize assembly time on a placement machine

This paper addresses a difficult problem that often arises in printed circuit card assembly systems: how should circuit cards be grouped into families to decrease total assembly time? Traditional approaches to this problem focus on setup time, without considering the possible impacts on processing time. A new approach for selecting card families to be processed on an assembly machine is developed to minimize total assembly time through the joint consideration of setup time and processing time. The overall approach for addressing the card grouping problem involves capturing the lower level machine configuration decisions through an empirical estimator function and incorporating this function with the higher level card grouping problem. The card grouping problem is solved using a branch-and-bound algorithm supplemented by techniques to improve the solution time. An industrial case study is conducted for a turret style placement machine, a Panasonic MV150 machine. The results demonstrate the positive impact of including the lower level decisions on the total assembly time and system throughput for certain types of assembly machines. In addition, this research provides insight on other process planning problems, such as line assignment and card sequencing, by demonstrating the importance of incorporating the related lower level decision problems.