‘Stochastic Clearing under Multiple Brownian Motion Input Processes’
by Bo Wei
Singapore University of Technology and Design
Join Zoom Meeting
Meeting ID: 337 069 8252
Stochastic clearing theory has wide-spread applications in the context of supply chain and service operations management. While the existing literature in the area has focused on the single input case, our focus is on the case of multiple input processes with the goal of investigating the structure of exact optimal clearing policies, directly applicable in the context of shipment consolidation practices, under both cost- and service-based performance measures. To this end, _rst, we demonstrate that the previously established cost-optimal quantity-based policy (QP) is not cost-wise superior to the cost-optimal time-based policy (TP): a notable result which is fundamentally di_erent than for the case of single input process. Next,we identify a new set of policies, referred as (TQ + T)-policies, and show that the cost-optimal (TQ+T)-policy is either the optimal QP or the optimal TP. Building on these results, we propose an instantaneous rate policy (IRP) and prove its optimality among a large class of renewal-type clearing policies, in terms of the average cost. In relation to service-based performance, we study the behaviour of average weighted delay rate (AWDR)under the IRP. We show that, for a _xed clearing frequency, the IRP also achieves the lowest AWDR among a large class of renewal-type clearing policies. Hence, our newly established IRP leads to a win-win operational strategy in terms of both cost- and service-based criteria. Our methodological contribution is in the delivery of a uni_ed method to calculate both cost- and service-based performance measures for a general class of renewal type clearing policies by developing novel martingale-based arguments. Our results are of practical value in the context of design and operation of shipment consolidation programs.
Bio: Bo Wei is a research fellow at Singapore University of Technology and Design from November 2020 onward. Prior to his current position,he was a research fellow at National University of Singapore since 2015.He obtained his PhD in Industrial Engineering from Texas A&M University in United States in December 2014 under the supervision of Drs.S_la C_ etinkaya and Daren B. H. Cline. He also holds an MS in Control Theory (2008) and a BS in Automation (2005), both from the University of Science and Technology of China. His research expertise and interests are in applied probability and large-scale constrained optimization with applications in supply chain and service systems. Wei’s publications in these areas have appeared in Operations Research Letters, Probability in Engineering and Informational Sciences, Annals of Operations Research, and Mathematical Methods of Operations Research.