This REU project focuses on developing a discrete-event simulation model using Simio to analyze and improve production performance at ACME Transmission Parts, a Tier 2 automotive supplier manufacturing torque converters and oil pans. The plant is currently struggling with low on-time delivery, high scrap, quality issues, and labor inefficiencies—all of which place the operation at risk of closure. Because torque converters generate higher revenue and exhibit significantly worse performance (60% on-time delivery, 0.50% return rate, major scrap losses, and heavy reliance on overtime), the simulation study will center on the torque converter value stream.

Using a provided business case, value stream maps, plant layouts, and improvement proposals, the student will construct a baseline Simio model that captures the current system constraints: long cycle times, weekly production runs, substantial changeover time, limited uptime, high WIP, and material-handling delays. The simulation will represent major process steps (i.e., component stamping, impeller/turbine/reactor assembly, final assembly, leak testing, and packaging) along with relevant resources such as labor, shared presses, forklifts, and overtime-dependent capacity.

Once the baseline system is validated against current operational metrics (throughput, labor productivity, WIP levels, lead time, and OTD), the student will experiment with a series of potential improvements drawn from the case materials. These include reducing changeover time, adding or redistributing labor across shifts, improving leak-test yield, increasing delivery frequency for materials, implementing pull scheduling, and altering batch sizes. Each scenario will be evaluated to understand the impact on throughput, WIP, cycle time, quality yield, resource utilization, overtime requirements, and the overall ability to achieve SQDC goals.

The goal of this REU project is to (1) provide an evidence-based improvement roadmap for ACME’s torque converter value stream and (2) give the student hands-on experience in data analysis, simulation modeling, experimentation, and interpreting manufacturing system performance. By the end of the project, the student will deliver a simulation model, a set of recommended improvements, and a quantitative comparison between current state performance and several proposed future states. This work will demonstrate how discrete-event simulation can guide operational decision-making in complex production environments and support data-driven continuous improvement.

Fig. 1: ACME Plant Layout.