Adaptive Bayesian Optimization Algorithm for Unpredictable Business Environments

This paper introduces an adaptive Bayesian optimization (BayesOpt) framework with dynamic conditioning and jitter mechanisms. The new framework enhances the adaptability and effectiveness of optimization in unpredictable business environments. The dynamic scaling in this framework dynamically modifies the mean objective function in each iteration, and adaptive conditioning functions. The adaptive acquisition jitter function enhances adaptability by adjusting the jitter of the acquisition function. The framework is tested using single-objective, multi-objective, and decoupled multi-objective functions. Statistical analyses which include t-statistics, p-values, and effect size measures (Cohen's d and Hedges g) reveal the superiority of the proposed framework over the original Bayes optimization. The primary contribution is developing a novel and effective optimization approach in stochastic environments, especially in the context of supply chain inventory management.