Process Improvement through Automated Feedback ( Preliminary Report )

Continuous quality improvement in a manufacturing environment is a major challenge. The environment is in a constant state of ux (machines break down, a consumable is bought from a new vendor, new processes or machines are introduced, etc). In our domain we are exploring the automated learning of process information. Our approach is to attack this problem simultaneously from three distinct sub elds. We are presently exploring machine learning, classical statistics, and neural networks as three possible ways of solving this problem. In this paper we will focus on the description of the machine learning task. We are at an early experimental stage of our research and therefore have only preliminary results.