Date of Graduation

12-2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Industrial Engineering

Advisor/Mentor

Manuel D. Rossetti

Committee Member

Edward A. Pohl

Second Committee Member

Nebil Buyurgan

Third Committee Member

Kevin M. Taaffe

Keywords

Applied sciences, Intermittent demand, Inventory, Lead time, Service level requirements

Abstract

This dissertation consists of three assays. The first assay examines the robustness of lead time demand models for the continuous review (r;Q) inventory policy. A number of classic distributions as well as distribution selection rules are examined under a wide variety of demand conditions. First, the models are compared to each other by assuming a known demand process and evaluating the errors associated with using a different model. Then, the models are examined using a large sample of simulated demand conditions. Approximation results of inventory performance measures - ready, rate excepted number of backorders and on-hand inventory levels are reported. Results indicate that distribution selection rules have great potential for modeling the lead time demand.

Incorporating distributions that preserve higher moment information into an inventory control system to determine the desired performance measures is a challenging task. One difficulty in applying such distributions is estimating the parameters from the data. In most cases only the demand per period is available. Thus, the demand per period moment data must be combined with the knowledge of the lead-times to represent the moments of the lead-time demand. The other difficulty lies in deriving closed form expressions that utilize an appropriate parameter fitting procedure. The second assay addresses these challenging issues by utilizing new parameter fitting strategies. The experiment results, collected under across a large number of simulated demand conditions, indicate that the models that preserve more flexible distributional form yield more accurate inventory performance measure results.

The focus of the third assay is to develop generic simulation optimization techniques based on sample average approximation (SAA) in order to set policy parameters of classical inventory systems having constrained service levels. This work introduces a policy optimization procedure for the continuous review (r;Q) inventory system having a ready rate service level constraint. Two types of SAA optimization procedures are constructed based on sampling from two different simulation methods: discrete-event and Monte-Carlo simulation. The efficiency of each sampling method is evaluated through a set of experiments under a compound Poisson demand process. In addition, the applicability of the proposed optimization procedure to the other re-order type inventory systems is discussed.

Download

Share

COinS