Uncertain Context Factors in ERP Project Estimation are an Asset: Insights from a Semi-Replication Case Study in a Financial Services Firm

This paper reports on the findings of a case study in a company in the financial services sector in which we replicated the use of a previously published approach to systematically balance the contextual uncertainties in the estimation of Enterprise Resource Planning (ERP) projects. The approach is based on using three techniques, a parametric model, namely COCOMO II, a portfolio management model, and Monte Carlo simulations. We investigated (i) whether the adjustment of uncertain cost drivers in the COCOMO II model increases the chance of project success in a portfolio of ERP projects, (ii) which cost drivers of the COCOMO II model can be adjusted in a way that maximized the chance of portfolio success under time constraints, and (iii) which cost drivers of the COCOMO II model can be adjusted in a way that maximized the chance of portfolio success under effort constraints. We found that 11 COCOMO II cost drivers can be changed so that the change impacts the project outcomes under both time and effort constraints. This result is different from the result in the first case study in which 13 such factors were found.