CCE Theses and Dissertations


A Simulation Model of Large-Scale Software Reuse Implementations: A System Dynamics Approach

Date of Award


Document Type


Degree Name

Doctor of Philosophy in Information Systems (DISS)


Graduate School of Computer and Information Sciences


Sumitra Mukherjee

Committee Member

Michael J. Laszlo

Committee Member

Lee Leitner


Large-scale software reuse, such as SAP's R13 Enterprise Resource Planning (ERP) system, has the potential to significantly reduce system implementation timelines. However, most ERP projects fail to deliver on time and within budget. While each implementation faces unique challenges and differing requirements coverage, the management decision to build custom programs instead of reusing the provided software introduces additional risk and effort to an already complex undertaking. The goal of this research is to evaluate the use of an original system dynamics model, combined with a novel approach to function point calculations, as a means to accurately predict the Realization Phase duration of SAP implementations.

To achieve this goal, a new function point counting approach to SAP transactions was combined with the development of a system dynamics model that simulates the Realization Phase of SAP Projects. Both the function point approach and dynamic model proved effective in replicating and predicting the behavior and timelines of completed projects across multiple firms. The first firm, Project Alpha, served as the reference baseline for model calibration (MAE=.09 R2=1). Data from a second firm, Project Beta, proved the portability of the model across different organizations (MAE=.07, R2=.98). The model's predictive performance was then tested against the project duration of a third firm. Project Gamma. Using the data from Project Gamma, the model estimated a project duration of 127 days versus the actual 134 days. Given the current level of estimating accuracy within ERP implementations, this research offers significant improvement to professional practice. Finally, the relationship between reuse and productivity was demonstrated by varying the modeled amount of reuse over several simulation runs. Each simulation provided insight on the dynamic relationship between the amount of reuse and project duration.

This document is currently not available here.

  Link to NovaCat