Depending on the software system to be developed, software development frameworks can yield different results depending on the complexity of the systems to be developed. While Plan-Drive or Sequential Software Development frameworks, e.g. Waterfall, have demonstrated their value developing simple systems; iterative or agile, e.g. Scrum, has done similarly with complex systems. Agile frameworks has demonstrated their success on small companies or start-ups, however, while scaling them to large-scale enterprises, the results are not as successful as expected. Deciding on the Software Delivery Life Cycle (SDLC) approach is an ambiguous challenge for teams working under the analysis of requirements for software development. Agile frameworks, such as Scrum, Kanban, or eXtreme Programming (XP), and more specifically agile techniques, fit quite well in those uncertain contexts and propose different perspectives in the referred frameworks that face quite a lot of challenges and cause friction when trying to apply them in the referred enterprises. These challenges and frictions can be caused by a variety of variables, such as enterprise culture, team maturity, individuals, or psychological safety, among others. The objective of this Ph.D. is to (1) synthesise existing empirical evidence via a Systematic Mapping Study, SMS, (2) use the results of this mapping study to propose a flexible and adaptative evidence-based framework for validating agile techniques in large-scale scenarios, and (3) empirically validate the framework in large-scale enterprises.

@inproceedings{paperbot11813,
  title = {Challenging Agile: Empirical method's tailoring validation at large-scale},
  author = {Francisco Gracia-Ahufinger},
  booktitle = {ICSE},
  year = {2026},
  doi = {10.1145/3774748.3787675}
}