OPTIMAL POLICIES FOR THE YIELD LEARNING-PROBLEM IN MANUFACTURING SYSTEMS

The ''yield learning'' problem encountered in many manufacturing syste ms involves controlling the production rate of a process so as to maxi mize cumulative rewards over some time interval. Because yield improve ment occurs as a result of discrete events coinciding with lot process ing completions, a basic tradeoff arises as high production rates incr ease rewards for a given yield but also cause longer average lot lead times which delay the events that trigger yield improvement. We show t hat in some cases simple bang-bang solutions can be obtained for this problem through standard optimal control techniques.