Enhancing Architecture Design Decisions Evolution with Group Decision Making Principles

Abstract

In order to build resilient systems, robust architectures are needed. The software architecture community clearly recognizes that robust architectures come from a robust decision-making process. The community also acknowledges that software architecture decision making is not an individual activity but a group process where architectural design decisions are made by groups of heterogeneous and dispersed stakeholders. The decision-making process is not just data driven, but also people driven, and group decision making methodologies have been studied from multiple perspectives (e.g., psychology, organizational behavior, economics) with the clear understanding that a poor-quality decision making process is more likely than a high-quality process leading to undesirable outcomes (including disastrous fiascoes).

In this work, we propose to explicitly include group decision making strategies into an architecting phase, so to clearly document not only the architectural decisions that may lead to the success or failure of a system, but also group decision making factors driving the way architecture design decisions are made. In this regard, this work defines a group design decision metamodel (for representing group design decisions and their relationships), together with ways to trace group design decisions towards other system life-cycle artifacts, and a change impact analysis engine for supporting evolving design decisions.