Abstract
This thesis has studied stochastic models and analysis of warehouse operations. After an overview of stochastic research in warehouse operations, we explore the following topics. Firstly, we search optimal batch sizes in a parallel-aisle warehouse with online order arrivals. We employ a sample path optimization and perturbation analysis algorithm to search the optimal batch size for a warehousing service provider, and a central finite difference algorithm to search the optimal batch sizes from the perspectives of customers and total systems. Secondly, we research a polling-based dynamic order picking system for online retailers. We build models to describe and analyze such systems via stochastic polling theory, find closed-form expressions for the order line waiting times, and apply polling-based picking to online retailers. We then present closed-form analytic expressions for pick rates of order picking bucket brigades systems in different storage profiles, and show how to combine storage policies and bucket brigades protocols to improve order picking productivity. Finally, we propose a new warehouse design approach oriented to improving revenue management of public storage warehouses. Our experiments show a proper facility design can significantly improve the expected revenue of public storage warehouses.