June 8, 2017 - June 8, 2017
3:30 pm – 4:30 pm
Sinem Özkan and Özgün Öztürk
Two visiting PhD students will present their work
A new production policy for M/M/s make-to-stock system and its renewal analysis
This study considers a production setting of a single-item, make-to-stock production system with parallel production channels, fixed start-up cost and lost sales. The decision should be addressed for managing such a system is the production control decision that dictates when to start production and how much to produce i.e., how many production channels should be activated. If the decision is to activate new channels, a setup cost is incurred per channel. The literature of the control of make-to-stock queues is extended by considering fixed system costs and multiple servers at the same time. We first develop the dynamic programming formulation of the problem and then characterize the structural properties of the optimal policy via value iteration algorithm. According to the numerical observations, optimal production policy is highly dynamic and does not have systematic behaviour for practical purposes. Hence, new/alternative policies that have well-defined structures and are easier to apply are proposed. The main contribution of this research is to calculate the average cost per unit time for M/M/s make-to-stock queues with fixed costs and lost sales under the proposed policy. It is obtained by conducting a renewal analysis. Finally, we get a very practical tool which calculates the average cost for M/M/s make-to-stock queue with setup costs in milliseconds. This study can be adapted for the systems with different service times such as Erlangian and Phase-type.
Sinem Özkan is a PhD student and research assistant at the Department of Industrial Engineering of Yaşar University in İzmir, Turkey. She received her B.S. and M.S. degrees in Industrial Engineering from Yaşar University in years 2012 and 2016, respectively. Before she joined the M.Sc. program, she worked one year as a production engineer in Hugo Boss Clothing Company. Then, she started working at the Department of Industrial Engineering of Yaşar University in 2013. Her research interests include production/inventory systems, stochastic modeling and optimal control.
Control of M/Coxian-2/s make-to-stock production systems
In this study, we consider a make-to-stock production environment with multiple processing channels, several customer classes, fixed production start-up costs and lost sales. Demands of customer classes are generated from independent Poisson processes. Processing times are assumed to be independent two-phase Coxian random variables. Each phase of Coxian distribution is an exponential random variable corresponding to a specific stage in production and there is a certain visiting probability from phase-one to phase-two. Phase-type processing times allow us to model the systems with rework/inspection operations. The problem is to control the production and allocate the on hand inventory among different customer classes. We extend the production-inventory control literature by considering phase-type production times, several customer classes, parallel production channels and start-up costs at the same time. The dynamic programming formulation of the problem is developed and optimal production and rationing policies are characterized under average system cost criterion. In addition to this, a dynamic rationing policy and several production policies are proposed and their performance analyses are carried out.
Özgün Öztürk is a research and teaching assistant at the Department of Industrial Engineering of Yasar University in Izmir, Turkey. She received her B.S. and M.S. degrees from the same department in 2014 and 2016, respectively. She is currently a PhD student and working on the control of production-inventory systems. In addition to dynamic programing and production-inventory control, her research interests also include stochastic processes and simulation. The courses that she has served as a teaching assistant are Engineering Statistics, Analytical Reasoning, and Discrete Computational Structures.
Location: TU/e, Building 62, Paviljoen F9. Feel free to join, if you are interested!
Eindhoven University of Technology