This research proposes a collaborative optimization strategy for revenue management and discount regulation that incorporates multiple uncertainty factors such as pricing, regional variation, bargaining, and margin fluctuations arising from sales activities.
To account for the dynamic variability of operational costs in uncertain environments, a reliability optimization model is established. In addition, a region-wise Kriging scheme is developed to construct a nonlinear reliability equivalence model. This design substantially reduces computational costs during both experimental design and the optimization process. Building on these estimations, the localized surrogate models are further embedded within a trust-region scheme. Periodic recalibration is used to preserve model fidelity and ensure the robustness of candidate solutions. A heuristic optimization algorithm, namely the Hiking Optimization Algorithm (HOA), is introduced to address this problem.
The case study demonstrates that the reliability-based design optimization model can more effectively resolve revenue management problems such as order sale uncertainty. Furthermore, the validity of the model, the efficiency of the algorithm, and the practicality of the collaborative optimization strategy are empirically verified. Sensitivity analysis further reveals that alternative planning strategies should be tailored to different optimization objectives, thereby strengthening decision-making robustness and reinforcing the framework's practical relevance.
The study provides theoretical guidance and analytical decision-support tools for joint decision-making in supply chain management. The framework is broadly applicable to uncertain supply chain contexts characterized by multi-sourcing, stochastic demand, and variable lead times.
