Risk preferences, newsvendor orders and supply chain coordination using the Mean-CVaR model

Abstract

The standard newsvendor model assumes that decision-makers in the supply chain are risk-neutral. This paper explores the newsvendor problem using risk measures building on the classical normative definition of risk preferences by certainty equivalents to model risk-averse, risk-neutral, and risk-taking decision-makers. We suggest the Mean-CVaR model based on the popular risk measures Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR). The objective function is characterized by a tolerance level differentiating between low and high profits and by a pessimism level representing the weight of low profits. For the first time, we prove that the optimal order quantity is higher (lower) than the risk-neutral order quantity if and only if the newsvendor is risk-taking (risk-averse). The equivalence of product-specific risk preferences and the size of the order quantity enables the determination of the Mean-CVaR parameters from given order quantities or from a target profit. Further, based on supply chain’s objective maximization criterion, we discuss supply chain coordination with Mean-CVaR decision-makers. We show that a simple wholesale price contract can coordinate the supplier-buyer supply chain and completely characterize the decision makers’ risk preferences under supply chain coordination. For example, if the coordinator (supplier) is risk-averse, a coordinating wholesale price exists for a less risk-averse buyer as well as for a risk-neutral and risk-taking buyer. We also determine the Mean-CVaR parameters if the coordinator specifies a target for the supply chain profit and the buyer communicates a target profit. In this case, the coordinating wholesale price depends on the ratio of the target profits and we show the impact on supply chain efficiency.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 22 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.