Maintenance/production planning with interactive feedback of product quality

This paper seeks to develop an optimal stochastic control model where interactive feedback consists of the quantity of flawless and defective products. The main objective of this study is to minimize the expected discounted overall cost due to maintenance activities, inventory holding and backlogs.

The model differs from similar research projects in that, instead of age‐dependent machine failure, it considers only defective products as feedback into the optimal model for maintenance and production planning. In this paper a near optimal control policy of the system through numerical techniques is obtained.

In this paper, a new model in which the system's retroaction is the quantity of defective products is presented, considering that defective products are a consequence of global manufacturing system deterioration. Instead of taking into account machine failure and human error separately, it considers a defect in product as being the consequence of a combined failure; this consideration allows one to be more realistic by merging all failure parameters into a single one. A new stochastic control model, which focuses on defective products, inventory, and backlog, has been developed.

This approach extended the concept of hedging point policy to the quantity of defective products combined with preventive and corrective maintenance strategies. The control policy obtained has a bang bang structure and is completely known for given parameters.

The integration of maintenance and production strategies has been mainly focused on the machine. Many research projects have been focusing on the age when dealing with machine failure. It is considered as the main target of the cost reduction in maintenance engineering departments. The originality of this paper is the taking into account of all operational failures into the same optimization model. It brings a value added to high level of maintenance and for operation managers who need to consider all failure parameters before taking decisions related to cost.