All manufacturing processes have variability.
How you handle variability can make it much easier to achieve your goals – from increased capacity and quality to decreased scrap and downtime.
Reducing the variability of key process parameters enables you to drive to your plant’s processing constraints and specification limits. There are a lot of ways to decrease variability — I’ve found that one of the most efficient and cost effective is through a Model Predictive Control (MPC) solution.
MPC makes your processes more stable – and predictable – so you can consistently get as close as possible to your targets and maintain optimal control.
There are four types of scalable analytics:
- Descriptive: What happened?
- Diagnostic: Why did it happen?
- Predictive: What will happen and when?
- Prescriptive: What should I do about it?
MPC is an example of a prescriptive analytic, correct for the process disturbance before it affects your product quality.
With MPC you build a mathematically defined multi-variable model built upon steady state and dynamic relationships. Process control utilizing the multi-variable model is able to predict future process values and coordinate setpoint changes for optimal performance while rejecting disturbances.