November 08, 2021

Srinivas

Spend more money on cutting tools to ensure you’re producing quality products on time? Or save on energy costs and slow production? Those are the types of trade-offs that manufactures make daily. Now, a Mizzou Engineer has devised a model to help inform those decisions and optimize manufacturing processes.

Sharan Srinivas, assistant professor in industrial and manufacturing systems engineering, developed a model specifically focused on job sequencing and replacement of cutting tools. He and co-author Mohamed Salama, a PhD student in industrial engineering, outlined the system in the October issue of Applied Soft Computing.

In manufacturing, cutting tools are used to extract material from components. For instance, manufacturing cell phone cases requires holes to be cut for charging devices, camera lenses and buttons.

Like kitchen knives, these cutting tools become dull over time. And similar to a dull steak knife, cutting tools don’t work as well when they’re not sharp.

“In a manufacturing setting, you have cutting tools that deteriorate quickly, sometimes within the same day,” Srinivas said. “And with gradual failure of the cutting tool, the quality of your finished product is likely to be nonconforming, and the product may not pass quality check. To avoid this, manufacturers prematurely replace tools.”

However, replacing tools prematurely is not only costly, it also takes time, which could delay production.

The speed of production is also a factor. Running cutting machines quickly can help meet production deadlines, but it also requires more energy. Slowing down production is a better, more energy efficient option if manufacturers aren’t in a hurry to finish a job.

Srinivas’s model takes all of these factors into account and provides suggestions on how to arrange jobs. With information such as deadlines, willingness to replace tools and types of products, the system finds the optimal job sequence.

What’s novel about the approach is that it uses a simulated annealing algorithm with an adaptive neighborhood search procedure. This essentially means the system randomly searches for a solution, then looks at related solutions to find the one that best meets the manufacturer’s needs.

Once the model determines the best solution, it provides a chart, showing the sequence in which the jobs should be processed and when a tool should be replaced. For instance, some jobs might need to be done first when the cutting tool is sharpest; some jobs can be slowed to save on energy costs; and other jobs might be best to complete just before the cutting tool is replaced.

“We’re trying to minimize the weighted sum of costs associated with due date violation, cutting tools utilized and energy consumption,” Srinivas said. “This system provides the best sequence of jobs and maintenance to achieve a low manufacturing cost.”