Submitted to manuscript Optimal Order Batching in Warehouse Management: A Data-Driven Robust Approach Vedat Bayram Department of Industrial Engineering, TED University, Ankara, Turkey, vedat.bayram@tedu.edu.tr Gohram Baloch Department of Management Sciences, University of Waterloo, Waterloo, Ontario, Canada, ggohram@uwaterloo.ca Fatma Gzara Department of Management Sciences, University of Waterloo, Waterloo, Ontario, Canada, fgzara@uwaterloo.ca Samir Elhedhli Department of Management Sciences, University of Waterloo, Waterloo, Ontario, Canada, elhedhli@uwaterloo.ca More than ever, the role of warehouses is crucial to supply chain efficiency. Optimizing warehouse processes has direct impact on supply chain responsiveness, timely order fulfillment and customer satisfaction. In this work, we focus on the picking process in warehouse management and study it from a data perspective. Using historical data from an industrial partner, we introduce, model, and study the Robust Order Batching Problem (ROBP) which groups orders into batches to minimize total order processing time taking into account uncertainty caused by system congestion and human behavior. We provide a generalizable, data- driven approach that overcomes warehouse-specific assumptions characterizing the majority of work in the literature. We analyze historical data to understand the processes in the warehouse, to predict processing times, and to improve order processing. For that purpose, we introduce the ROBP and develop an efficient branch and price algorithm based on simultaneous column and row generation. The algorithm is embedded with alternative prediction models such as linear regression and random forest to predict processing time of a batch. We conduct extensive computational experiments to test the performance of the proposed approach and to derive managerial insights based on real data. We found that forming batches at full capacity is not always optimal, and that the congestion level and level of conservatism have a significant impact on batching decisions. The data-driven prescriptive analytics tool we propose achieves savings of 7-8 minutes per order, which translates into a 14.8% increase in daily picking operations capacity of the warehouse. Key words : Warehouse Management; Order Batching; Batch Processing Time Prediction; Data Analytics; Robust Optimization; Branch and Price; Simultaneous Column and Row Generation; Data-Driven Prescriptive Analytics History : 1. Introduction The role of warehouses in supply chains has increased over the years. With the rapid increase in online shopping and the advances in automated storage and retrieval systems, warehouses have 1