A newsvendor analysis of a binomial yield production process

Abstract

We study a binomial yield production process using a newsvendor approach by considering the mismatch costs between a given demand and a non-defective yield amount. Using a normal approximation, we reformulate the original discrete and exact model as a continuous and approximate model. We conduct a comparative static analysis of the parameters in the approximate model and derive the monotone properties of the (approximate) optimal solution in the case that the given demand is sufficiently large. The analytical results are all consistent with our insights and are also supported by economic explanations. Our numerical study with sample-based optimization indicates that the approximate model is sufficiently close to the exact model in most real-world examples, with some limiting cases included. Then, a simple and naive solution is found to have significant suboptimality. Finally, the results of a sensitivity analysis of the model parameters are confirmed numerically.