TABLE OF CONTENTS In the ever-changing e-business environment, project design, delivery, and management is usually a highly complex and risky process due to the very dynamic and uncertain nature of business requirements, customer expectations, and technology, among other complexity drivers. Therefore, project performance and health tracking, prediction, and management require adaptive decision support tools that integrate various business data and suggest appropriate response actions in a sense-and-respond manner (Haeckel, 1999). To track and manage project health effectively, one needs to collect and analyze up-to-date information about various aspects of a project and identify possible deviations (trends) from initial and/or expected plans. Advanced data analysis techniques can provide valuable input for informed prediction of future project health by identifying emerging trends (patterns). The project manager can then respond proactively to early warnings in the project life cycle.
However, to design such adaptive management systems, a number of requirements have to be met. First, one must gather and consolidate as much historical data as possible on various aspects of current and previous projects. Then, using data analysis models and domain knowledge provided by subject matter experts, one has to derive appropriate process metrics and rules to serve as a baseline for:
Quantification of project quality and performance
Identification of emerging trends
Prediction of future project health and financials
In a typical system integration project scenario for example (see Figure 7.1), such metrics and rules would be determined by a combination of several factors stemming from the project financials,...
