Order picking – the retrieval of units from their storage locations to fill customer orders – is one of the highest priority areas in warehousing to meet both customer and cost expectations. Selecting the most suiting order picking strategy is a complex combinatorial problem, with research very scarce. Traditionally, once selected, the strategy stays static, even though it may become invalid due to unexpected events. We build seven well-known order picking strategies in simulation software to study the effects of variability using multiple performance measures. Based on this, we discuss a concept using machine learning to develop a real-time response model in daily operations decision-making. We focus on manual order picking operations, as this is the large majority of warehouses still today.
Katja Linda Thamm joined Kühne Logistics University as PhD candidate in July 2019 supervised by Prof. Dr. Sandra Transchel and Prof. Dr. Henrik Leopold. In her research, Katja combines actual observations in reality with operations management and machine learning, specifically in manual order picking warehouse operations. Katja has professional experience since 2007 in SCM and logistics, focusing on optimization and digitalization. Since 2017 she works as consultant with Retail Capital Partners AG, who is also project sponsor. She received her MSc in Global Logistics at KLU in 2012, with being KLU Master’s Thesis Award winner. In 2007, she received a Diploma in Business Administration, specialized in Forwarding, Transport and Logistics at DHBW Mannheim. Katja is now in her second year of the PhD program after maternity leave.