A machine learning approach to dynamic pricing in multi-channel transportation supply chains

Effective pricing strategies are critical for optimizing revenue and maintaining competitiveness in transportation supply chains, particularly in multi-channel environments. This paper presents a machine learning-driven dynamic pricing approach designed to optimize ticket prices across multiple transportation modes, service classes, and sales channels. In particular, the proposed approach integrates predictive analytics and optimization techniques to estimate customer demand, price sensitivity, and revenue potential while accounting for operational constraints such as capacity limits and market share requirements. Machine learning enables the optimization model to dynamically adjust pricing strategies based on historical demand patterns and real-time market fluctuations. To demonstrate its applicability, the approach is applied in a case study in the transportation sector, illustrating its role in optimizing pricing decisions. Additionally, sensitivity analysis highlights the model's robustness against capacity changes, demand fluctuations, and pricing constraints. The findings emphasize the role of supply chain analytics in enhancing pricing strategies, making them more adaptive, data-driven, and resilient to market dynamics.