by Ni, Y.; Werner, F. .
Series: 2017-05, Preprints
The improvement of the performance of material handling tools (MHTs) and the work in process (WIP) in a discrete manufacturing system have a great importance for increasing the efficiency of the production. To this end, the static and dynamic status of MHTs are analyzed in this paper. A Markov decision process (MDP) is used to model the MHT problems. The quantified relationships between MHTs and WIP will be discussed within the CONWIP (constant WIP) and Little’s law methodologies. A dynamic programming (DP) based algorithm is developed to determine a solution for the MDP model. To reduce the computational complexity of the DP algorithm, an appropriate modification is introduced. Computational experiments are conducted in a discrete semiconductor factory and the proposed MDP+DP method is compared with simulation. The computational results show that the developed method produces rather good feasible solutions.
Material handling tool; Markov decision process; Dynamic programming; Discrete manufacturing