Supply Process Improvement Decisions for a Newsvendor with Random Capacity

Abstract This paper considers a firm with random capacity that makes decisions on supply process improvement in a newsvendor setting. We find that the firm benefits from increased mean capacity, from reduced variance in capacity, and from monotonicity in the increasing concave order of capacity, despite diminishing returns. Moreover, we identify positive moderating effects on process improvements of profit margin and demand magnitude. When the capacity follows specific distributions whose mean and variance are jointly parameterized, we investigate comparative statics via analytical derivation and numerical experiments. The results are consistent with those obtained under a general distribution, where process improvement is more effective when initiated earlier rather than later. Analyzing various capacity distributions reveals that the random capacity's optimal parameters related to its mean increase with the firm's profit margin. However, the associations between these parameters and demand variability are subtle and may change under different procurement costs. The capacity's optimal parameters related to its variance decrease with higher profit margins but increase with greater demand variability. Our findings yield several managerial insights.