Sponsor:

Avaya Inc.

Research Team:

Scott J. Mason, Michael H. Cole, Bashyam Ramasami

Universities Involved:

University of Arkansas

Start Date:

09/01/01

End Date:

08/31/02

Summary:

This project’s objective is to investigate and develop improved inventory approaches for a complex, multi-stage, in sourced and outsourced production distribution environment at Avaya Inc. This is segmented in many ways depending on time available, data available, and resources available.
To help Avaya’s material planning team understand the material planning function, an MS-Excel emulator was created.
Avaya’s stated desire was to streamline current inventory management techniques through a better understanding of the existing SAP-based ERP system. The focus of the project was to help the members of Avaya’s material planning team to comprehend the material planning function that operates under the ERP system.

The primary deliverable of this research engagement was the development a Microsoft Excel®- based emulator of Avaya’s inventory management system.
Avaya Inc. has three levels of MRP decision-making: Coaches/Managers, Forecasters and Planners. Each level makes decisions that act as input to the next lower level.
By emulating the system, Avaya’s material planners now better associate with the original system and are more able to learn its associated complexities.

The emulator performs three main tasks:

1. It enables planners to find the optimal ROC and LSK combination for a particular demand forecast.
2. It enables forecasters to compare their forecasts to historical demand and to smoothed forecasts.
3. It enables managers to evaluate planners and forecasts according to inventory performance.

The emulator addresses the issue of the bounded rationality of the users by emulating just the material planning function from within the organization-wide ERP system.

There is a better understanding of the impact of forecasting on the order quantities and order efficiencies.

The tool helps the planner to observe that the ROC should be a function of lead-time and the demand.