<p>With the increasing complexity of the supply chain, the operations related to warehouses have become more critical. When warehouse performance is evaluated, many criteria are to be considered. Since multiple operational activities occur in a warehouse, it might be difficult for a manager to identify the critical criteria set that will have a pivotal role in improving warehouse performance. In this paper, we address the following research questions: What are the critical criteria warehouse managers should focus, and what is the methodology that can be used to identify these critical criteria? In this study, we develop a machine learning based hybrid methodology to identify the critical ones. Thematic analysis is used to organize and group the criteria (one hundred and seven) into different themes. The criteria set is grouped and organised using the density-based spatial clustering algorithm. The criteria within each cluster and the clusters are then ranked by using PROMETHEE II method. A group decision making approach is used in order to minimize the decision maker’s bias. It was seen that routing related criteria followed by inventory sizing and warehouse design related criteria were vital ones. After ranking criteria, it becomes easier for the managers to identify those critical criteria, which could impact the most on warehouse performance.</p>