Risk averse supply portfolio selection with supply, demand and spot market volatility

• Published in: Omega (Oxford) Vol. 57; pp. 40 - 53
• Main Author: Merzifonluoglu, Yasemin
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.12.2015; Pergamon Press Inc
• Subjects: Conditional value-at-risk; Enterprise risk management; Market prices; Mathematical models; Procurement management; Random variables; Securities markets; Spot markets; Studies; Supply chain management; Supply disruptions; Supply disruptions; Conditional value-at-risk; Procurement management; Supply chain management; Spot markets
• ISSN: 0305-0483; 1873-5274
• DOI: 10.1016/j.omega.2015.03.006

Enterprise Risk Management (ERM) has become one of the most essential subjects in business management. This paper establishes how risk modeling can be applied to supply chain management, specifically to supply portfolio procurement decisions of a firm. In a single period setting, parts can be procured via traditional forward contracts, option contracts or spot purchases. Customer demand and spot prices are random and possibly correlated and firm׳s primary suppliers are subject to complete disruptions and yield uncertainties. This paper analyzes several scenarios where the spot market is not available, available for buying only, and available for both buying and selling. This article develops and solves mathematical models considering the risk neutral and risk averse (CVaR) objectives independently or simultaneously. For the special case of normally distributed random variables and a risk neutral objective, optimality properties were developed. A broad numerical study examines the sensitivity of procurement strategies to key problem parameters such as, risk attitude, demand and spot price volatilities, correlation between demand and spot prices and terms of option contracts. •We developed optimization models for procurement portfolio of a firm.•Risk neutral and risk averse objectives (CVaR) are considered.•Scenario-based LP models can be easily solved using commercial software.•A sample average approximation method (SAA) is proposed to approximate solutions.•Optimality properties are provided for the special case.