Marcel ILIE1, Augustin SEMENESCU2
Abstract. Supply chain systems are increasingly exposed to uncertainty due to fluctuating demand, variable lead times, and frequent disruptions across global networks. Conventional deterministic and frequentist models often fail to adequately address these complexities, as they lack mechanisms to incorporate prior knowledge and update beliefs dynamically. This study develops a Bayesian inference framework for supply chain management that integrates probabilistic modeling, learning, and optimization within a coherent decision-making structure. The framework employs hierarchical Bayesian models to capture multi-echelon dependencies and utilizes Markov Chain Monte Carlo (MCMC) techniques for posterior estimation and parameter calibration. By continuously updating prior distributions with new data, the model supports adaptive demand forecasting, real-time inventory optimization, and uncertainty quantification. Numerical experiments using simulated and empirical datasets demonstrate that the proposed Bayesian approach yields higher predictive accuracy, improved service levels, and enhanced resilience compared with traditional stochastic optimization methods. The findings highlight the capability of Bayesian inference to provide a flexible and data-driven foundation for robust decision support in complex and uncertain supply chain environments.
Keywords: Bayesian inference; stochastic modeling; supply chain management; uncertainty quantification; hierarchical models; MCMC; adaptive decision-making; resilience.
DOI 10.56082/annalsarscieng.2025.2.62
1 Associate. Prof. Ph.D. Georgia Southern University, 1332 Southern Dr. Statesboro GA 30458, USA, *Corresponding author:, milie@georgiasouthern.edu
2 Prof. National Science and Technology University Politehnica Bucharest, Spl.Independentei 313, Bucharest, Romania, augustin.semenescu@upb.ro
PUBLISHED in Annals of the Academy of Romanian Scientists Series on Engineering, Volume 17, No2