Document Type

Closed Project

Publication Date

Summer 2004


Stephen Singham

Course Title

Capstone Project

Course Number

EMGT 589/689


Inventory control -- Mathematical models, Business logistics, Delivery of goods -- Management


This study will present an inventory management model created with Arena and the efforts to develop and improve the model. In the first part, I tried to implement an Inventory Management model for a store for just one product. The work to build the model, the results and my endeavor to develop and improve the model are reported. In the second part, the model is improved to multiple products stage and implemented to three products level. An “Inventory Management Model” can really be complicated if we want to model all the real world complications. Therefore, the scope is tried to be kept as simple as possible. To do this, I have made several assumptions that are described through this report.

The initial intention was building a more complex and somewhat intelligent model. I was planning to use the Arena software work as an inventory optimization program, give the optimum figures for Replenishment Policy for when to order and how much to order in addition to showing the bottlenecks in the system and places to improve. However, as I studied the Arena software help files and researched over the internet, I could not find a way to implement this idea. Therefore I decided to go with a given “Replenishment Policy”, i.e. a managerial decision has already been made to identify when to order and how much to order. Maybe this given policy is not the best of possible. Through this project, this policy is tried to be improved too. Other than that, the bottlenecks and the limitations in the system are tried to be found and the system is improved to omit these deficiencies too.

Although the initial purpose was running the model with real data upon implementation, neither time nor my limited knowledge about the Arena software let me to do that. Therefore the model is tested with data generated by Arena software with random distribution functions. However, this does not affect the reliability of the model as seen through the report.


This project is only available to students, staff, and faculty of Portland State University

Persistent Identifier