EURO 2025 Leeds
Abstract Submission

2656. A simulation-driven order dispatcher algorithm for international orders in Trendyol

Invited abstract in session MD-47: Retail Supply Chain Management, stream Retail Operations.

Monday, 14:30-16:00
Room: Parkinson B08

Authors (first author is the speaker)

1. Mehmet Orhan YILDIRIM
2. MERT PARÇAOĞLU
3. Betül Ahat
Data Science, Trendyol
4. Ahmet Çınar
Data Science, Trendyol
5. Meltem Sanisoğlu

Abstract

Efficient international order fulfillment is crucial for e-commerce platforms managing multi-warehouse operations. This study introduces a dispatcher algorithm for Trendyol, engineered to minimize operational costs while satisfying diverse constraints inherent in international logistics. The algorithm considers various business requirements, including item demand, order origin, stock availability, warehouse prioritization, and customs regulations. It optimizes the total operational cost of an order by balancing warehouse expenses and shipping costs while determining whether to consolidate multiple packages. By applying advanced optimization techniques, it evaluates all feasible stock combinations based on predefined cost parameters, selecting the most cost-effective dispatch strategy. Additionally, it ensures efficient stock allocation across global storefronts by integrating warehouse priority mapping, which streamlines package consolidation. To further assess the impact of different policies on dispatch decisions, we developed a simulation framework that measures key business metrics such as unit shares of different warehouses and profit margins. The algorithm results show 5% saving in total operational costs while minimizing the total delivery/order through consolidation. This comprehensive approach enhances fulfillment efficiency, reduces logistic costs, and improves cross-border order accuracy, making it a scalable solution for high-volume e-commerce operations.

Keywords

Status: accepted

Back to the list of papers