Empirical studies clearly show that due to their inherent inflexibility and socio-technical inappropriateness, prescribed methods are rarely used or followed rigorously in software development projects. This fact is still recognized as one of the key reasons for failures in IT projects and more importantly, as a contributor to low quality software.

The main objective of the project is to circumvent the problems that hinder the use of software development methods in software development practice and by this to contribute to a higher success rate of IT projects and further to a higher quality of software.

Following this objective, our goal is to develop an innovative approach called intelligent Agile Method Framework (iAMF) that will simplify the use of the SME principles by only marginal user involvement (Goal1). iAMF will help us to capture new evidence from real cases giving us an important insight into the software development practice (Goal2).

The collection of data will be performed in two ways. For those organizations that use workflow-based systems to enact their software development processes, the data will be collected directly - from the logs, which record user activities. Unfortunately, not many software companies use computerized systems to support the enactment of their software development processes. For such organization we will develop a computerized system for semi-automatic knowledge acquisition through monitoring of the in-action software development methods. The main idea of the system is to monitor inaction methods and based on that:

a) To infer what organizations base methods are and thus to get real picture of what developers and other project members are doing on projects (base methods are formalized descriptions of typical ways of performing IT projects in an organization);

b) To supervise project performance with identification of any from base methods.

c) To keep based methods up-to date with capturing the knowledge gained on projects (by collecting and justification the reasons for declinations);

d) To guide developers and help them to follow base methods more rigorously.