The problem
Where the time was going
When an aircraft needs a vital part or a hospital needs a replacement for a knee or hip, the answer often starts at a casting company. This producer is one of the largest in the United States, making aircraft turbines, impellers, industrial turbines, prosthetic implants, airframes, and other components for the aerospace and medical industries, with casting and forging plants across the United States and Europe and several billion dollars in annual sales.
With many customers waiting at once, the castings facility could not project ahead how many resources it would need, when it would need them, or when a given job would finish. That uncertainty was hard on the business and its customers alike. Limited capacity let inventory pile up at times, while sudden demand for certain parts forced a scramble at others.
The stakes made the timing unforgiving. A single order might be a small part bound for an expensive engine on an even more expensive airplane, due within a set window or contract penalties would apply, and the strain pushed employees into long extra hours. The facility wanted to serve customers better, reduce that overtime, and free floor space by cutting the inventory sitting on the manufacturing floor.
What we modeled
Mapping the process, then testing the fix
The facility first looked at large corporate prediction-and-analysis systems, but the price tags and the maintenance and corporate overhead they demanded sent the team looking elsewhere. ProcessModel offered the same kind of insight without that weight.
Once the data was entered, the model identified bottlenecks and gave feedback that pinpointed where to improve. It revealed a constraint in the mold-making portion of the casting process, where inventory backed up, and it showed why excessive inventory built up and let the team test solutions against it before changing the floor.
The result
The proof, and the payoff
The model gave the facility a way to meet aggressive customer schedules with fewer late orders and to predict outcomes ahead of time instead of guessing. By correcting the inefficiencies it surfaced, the team could hold to its inventory-reduction plans and project future needs in advance, from capital equipment to staffing to process changes, which matters when investment casting can require an expensive six-to-twenty-month commitment in tooling and process development per customer.
Most of all, the producer could now manage what had been a juggling act. It regularly had twenty to thirty customer companies wanting orders filled in overlapping windows while it pursued new business, and the model let it organize schedules, tell each customer exactly when and how an order would be met, and keep reducing inventory and lead time at the same time.
Part of our work in manufacturing.