Investigating the Effect of Customization on Rework in a Higher Education Enterprise Resource Planning (ERP) Post-Implementation Environment: A System Dynamics Approach

Abstract

Management of enterprise resource planning (ERP) projects is exceedingly challenging due to their large scope and complexity. Schedule and cost overruns have been widely reported in ERP implementation for years. Not only are resources critical to successful ERP implementation, but the post-implementation requires substantial resources as well. However, the resources required to maintain and extend ERP systems after implementation are not well understood, especially in the presence of software customizations which increase the total cost of ownership. This research applies system dynamics simulation modeling, focusing on the rework cycle, to data of a case study toward three primary goals. First, it contributes to both the system dynamics literature and the pre-packaged software literature by exploring the post-adoption ERP environment and extending the task rework cycle. Secondly, it develops a simulation model, grounded in ERP literature and in well-established system dynamics project models, helping practitioners and organizations in making more informed customization decisions by giving them a tool to better assess the long-term resource implications of ERP customization. Lastly, it addresses research gaps recently identified regarding issues such as customization, project evaluation, effort estimation, impact of adoption, system maintenance, performing updates with enterprise apps, and potential reasons for system abandonment.

ACKNOWLEDGMENTS

The author would like to thank Dr. Ira Goldstein, Dr. Jack Rivituso, and the three anonymous reviewers for their thoughtful feedback and help in improving this article.

Notes

¹ Of the rework tasks completed for the June 2008 software upgrade, 94% were customization-related tasks.

² Vensim simulation software is used for developing, analyzing, and packaging high quality dynamic feedback models. The current implementation of molecules is an add-on to the Vensim software. Source: http://www.vensim.com/molecule.html

³ Pulse Train is a Vensim built-in function whose arguments are start time, duration, repeat time, and final time. This function allows for the control of bundle and upgrade releases at regular points in time.

Additional information

Notes on contributors

Meg Fryling

Meg Fryling is an assistant professor of computer science at Siena College and a former associate director of information technology services at the University at Albany (UAlbany). She received her B.S. in computer science from Siena College followed by her M.B.A. and Ph.D. in information science from UAlbany. During her professional tenure at UAlbany, Dr. Fryling spent more than 10 years managing the implementation, maintenance, and upgrade of information systems, including the university’s ERP system. Her research interests include IS acceptance and success, system dynamics (software project management), enterprise-wide information system (e.g., ERP) implementation and maintenance, and the dynamics of technology and society (e.g., cyberbullying).