Factory Simulation Optimization
Purdue University
As the facility layout sub-team lead, we were tasked with designing a new layout for a factory that produces 10 unique parts on 25 machines. Four categories of changes were tested: facility layout, ordering policy, job-selection, and machine selection. We tested each alternative solution in Arena Simulation Software.
-
Facility Layout: Grouped machines into work cells by grouping machines by part, ergo reducing transport times and setup-times.
-
Ordering Policy: Through nonlinear optimization, an optimal reorder point and reorder quantity was found.
-
Job-Selection: Optimized the queue rankings through implementing a hybrid selection rule minimizing the shortest remaining production time and setup time.
-
Machine-Selection: Implemented a hybrid selection rule that minimizes the queue length, setup time, and shortest travel time.
The work in process (WIP) reduced by 26% and the time in system decreased by 42%.