Designing Efficient Supply Chains with Unreliable Suppliers

In this research, a novel two-stage stochastic model is developed to find the optimal locations of the potential manufacturing facilities at the first stage, and to determine the optimal production levels, inventory levels, and quantities of raw materials and finished goods shipped among the members of the supply chain network at the second stage. The main feature of this research is that some suppliers are unreliable, and each unreliable supplier’s lifetime is assumed to be an exponentially distributed random variable. However, once the supplier breaks down or loses its ability to supply raw materials, it can be repaired. The length of time the supplier is out of service or does not operate is assumed to be another exponentially distributed random variable. The transitions between the two states of available when the supplier normally operates and unavailable when it is out of service follow a continuous-time Markov chain. Demands at markets are also random variables following normal distributions. The goal is to minimize the sum of first stage construction costs and the expected second stage production, inventory and shipping costs over the planning horizon while meeting service levels at the markets.