This paper presents a robust optimization model for integrated production-distribution planning in a three-echelon supply chain network comprising suppliers, manufacturers, and distributors. Using the Bertsimas-Sim robust optimization approach, we address uncertainty in demand and transportation costs across four transportation modes (road, rail, maritime, and air) to maximize net profit. The proposed Mixed-Integer Nonlinear Programming (MINLP) model is solved using GAMS/BARON 12.6 optimization software for four uncertainty levels (ρ=0.2,0.4,0.6,0.8) across five randomly generated scenarios. Comparative analysis between deterministic and robust models reveals that the robust model reduces standard deviation by 62-65% while maintaining only 0.75-1.79% reduction in average profit. Results demonstrate that the robust model achieves superior risk-return ratio, improving the coefficient of variation by over 60% across all uncertainty levels, making it particularly effective in high-uncertainty business environments.This research demonstrated that robust optimization provides a powerful and practical framework for integrated production-distribution planning under uncertainty. The Bertsimas-Sim approach successfully balances conservatism and optimality, achieving substantial risk reduction (62-65%) with minimal profit sacrifice (<2%).

The key insight is that robustness is not expensive – organizations can significantly enhance operational stability without compromising profitability. In an era of increasing supply chain disruptions, market volatility, and global uncertainty, robust optimization represents a critical tool for resilient supply chain management.

The proposed model and methodology are implementation-ready and can be directly applied by practitioners using standard optimization software. As uncertainty continues to characterize modern business environments, the adoption of robust optimization approaches will become increasingly essential for competitive advantage and long-term sustainability.