Title :
Coordinated quality control in a two-stage system
Author :
Yao, David D. ; Zheng, Shaohui
Author_Institution :
Dept. of Ind. Eng. & Oper. Res., Columbia Univ., New York, NY, USA
fDate :
6/1/1999 12:00:00 AM
Abstract :
The authors study a quality control problem in a two-stage system. They assume that at each stage units are processed in batches, and the rates are random variables with known distributions. Final products are supplied to customers under warranties or service contracts, with penalty costs associated with defective units. The focus is on coordinating the inspection procedures at the two stages. Using a stochastic dynamic programming approach, the authors show that the optimal policy at stage 1 is characterized by a sequence of thresholds, and at stage 2, by a priority structure, as well as a threshold structure. The key to optimality is a so-called K-submodularity property, which is a strengthening of the usual notion of submodularity
Keywords :
batch processing (industrial); dynamic programming; inspection; quality control; stochastic programming; K-submodularity property; coordinated quality control; defective units; inspection procedures; optimal policy; penalty costs; priority structure; random variables; service contracts; stochastic dynamic programming approach; threshold structure; two-stage system; warranties; Contracts; Cost function; Dynamic programming; Industrial engineering; Inspection; Marine vehicles; Quality control; Random variables; Stochastic processes; Warranties;
Journal_Title :
Automatic Control, IEEE Transactions on