This paper outlines experiences with using economic criteria to make architecture design decisions. It briefly describes the CBAM (Cost Benefit Analysis Method) framework applied to estimate the value of architectural strategies in a NASA project, the ECS. This paper describes the practical difficulties and experiences in applying the method to a large realworld system. It concludes with some lessons learned from the experience.

The benefits of a software system are assessable only relative to the business goals the system has been developed to serve. In turn, these benefits result from interactions between the system's functionality and its quality attributes (such as performance, reliability and security). Its quality attributes are, in most cases, dictated by its architectural design decisions. Therefore, we argue that the software architecture is the crucial artifact to study in making design tradeoffs and in performing cost-benefit analyses. A substantial part of such an analysis is in determining the level of uncertainty with which we estimate both costs and benefits. We offer an architecture-centric approach to the economic modeling of software design decision making called CBAM (Cost Benefit Analysis Method), in which costs and benefits are traded off with system quality attributes. We present the CBAM, the early results from applying this method in a large-scale case study, and discuss the application of more sophisticated economic models to software decision making.