In today's world of globalization and accelerated economic processes, logistics supply chains play a key role in ensuring business efficiency. Transportation costs, which account for 40-60% of the value of goods in the supply chain, can be significantly reduced by optimizing shipping routes, minimizing delivery times, reducing fuel consumption, and increasing profitability. This article analyzes route optimization strategies in logistics systems, with a focus on reducing transportation costs. The objective of the study is to develop a route optimization model based on artificial intelligence algorithms, adapted to real-world supply chain conditions. To achieve this goal, the following objectives are addressed: a review of theoretical approaches, the development of a methodology using a genetic algorithm and geographic information systems, case studies using examples of companies supplying consumer goods, and an assessment of the economic impact. The relevance of this topic is driven by the growth of online commerce, where timely delivery is critical for competitiveness. In Russia, with its vast territory and seasonal fluctuations in demand, this problem is particularly acute in regions with a well-developed supply network. The theoretical framework is based on classical optimization models, supplemented by machine learning methods and the concept of "green logistics" to reduce the carbon footprint. The methodology includes data collection, modeling with a cost-minimization objective function, and analysis of real-world cases. The practical significance of the study lies in the model's applicability to retail, manufacturing, and distribution companies, where a 15-25% cost reduction directly impacts profits. Recommendations include integration into ERP systems and phased implementation.