Abstract
Production systems with storages have recently attracted considerable attention from the reliability research community. The existing models assume a single type of resource consumed for product generation. This work extends the state of the art by considering a more general and practical model, where multiple resource supply subsystems (RSS) provide different kinds of necessary resources during the production cycle. Each RSS has a storage with limited capacity to save surplus resource, which may be supplied to the production unit when the RSS fails. The product unit may operate with different productivity levels, and RSS and storages can be chosen from multiple types. We formulate a new optimization problem, which chooses the productivity level of the production unit and the types of RSS and storage for each kind of resource required for the production to maximize the expected manufacturer's profit. A new numerical algorithm is proposed to evaluate the expected profit and the genetic algorithm is implemented to solve the proposed optimization problem. A detailed case study of a chemical reactor system with four RSSs supplying two reagents, catalyzer, and cooling water is provided to demonstrate the proposed model and influences of several model parameters on the optimization solutions.