Mitigating the risks of supply disruption under stochastic demand

This paper investigates the problem of sourcing decisions under stochastic demand considering quantity discounts and suppliers' failure risks. A Mixed Integer Non-Linear Programming (MINLP) model is developed considering different failure probability, capacity, price discounts, compensation potential, and transportation cost under a stochastic demand environment to determine the optimal suppliers' portfolio (i.e. the number of suppliers and order quantity allocation) for a single product in a single period. The model considers that each supplier has different failure probability, capacity, transportation cost, price discounts and compensations potential. The MINLP model is NP-hard in nature and difficult to solve with exact methods; therefore, a Real Coded Genetic Algorithm (RCGA) is employed here to solve it. Further, RCGA performance is benchmarked by comparing its results with the MINLP solver CitationBONMIN (Basic Open-source Mixed Integer Non-Linear Programming). A numerical study is conducted to illustrate the proposed model, a genetic algorithm and BONMIN to find the solution. Finally, different managerial implications have been drawn based on sensitivity analysis.