Agile methods have emerged largely from practice as a response to cumbersome managerial controls, which sought to address the so-called ‘software crisis.’ While the agile approach has gained widespread currency and appears to offer success, there are several weaknesses that can only be addressed through rigorous research on the topic. There is an inadequate theoretical and conceptual foundation of agile methods. For example, there are many flavours of agile, which are difficult to compare, but a central tenet of many is that agile practices are not indivisible but have to be applied in their entirety. Empirical evidence does not however support this assertion.
This research focuses on two key research areas:
- Lean practices from the manufacturing industry are seen by industry as potential solution to the ‘soft’ perception of agile methods. Lean practices bring a focus on metrics, quality management, root cause analysis and evidence-based process improvement. The research challenge here is adapting, implementing and validating these lean practices in a software engineering context.
- There is an inadequate rigorous evidential base to answer questions as to the effect of agile approaches on software cost, schedule, and quality. Establishing such a base is a major research undertaking but is absolutely critical in the context of establishing the latter in the context of evolving critical systems. Sound management and decision-making underpins the success of ECS projects. However, little research has focused on these issues. This gap becomes even more pertinent when we consider that contemporary implementations of agile go beyond small co-located teams as was originally intended, with non-standard implementations now widespread i.e. large teams, start-ups, distributed development environments, greenfield sites, educational environments, open source development, outsourcing, and systems maintenance. This presents new and different challenges for decision making, control, and the scaling of agile and lean. The team, rather than a person, manages the project, so issues such as customer involvement, shifting responsibilities and simultaneous development and testing require a new approach to project management and decision making.
The group is also looking at issues in a multi-project environment. This involves looking at the challenges of managing portfolios of software projects. This will include researching how to make portfolios more dynamic and identifying ways to bridge the gap between strategic intent and self-organising agile teams. – See more at: http://www.lero.ie/project/alps#sthash.yaXDrJeh.dpuf