Reinforcement learning based scheduling in semiconductor final testing

Semiconductor test scheduling problem is a variation of reentrant unrelated parallel machine problem considering multiple resources constraints, intricate {product, tester, kit, component} eligibility constraints, and sequence-dependant setup times, etc. A multi-step reinforcement learning (RL) algorithm called Sarsa(λ,k) is proposed and applied to deal with it. Allowing enabler reconfiguration, the capacity of the test facility is expanded and scheduling optimization is performed at the component level. In order to apply Sarsa(λ,k), the scheduling problem is transformed into an RL problem by defining state representation, constructing actions and the reward function. Experiments show that Sarsa(λ,k) outperforms the scheduling method in industry and validate the effectiveness of Sarsa(λ,k) to solve the scheduling problem.