A value-based approach to predicting system properties from design

Traditional engineering requires evaluating designs before implementing them. Evaluating a design predicts the properties of a reasonable implementation and the value of these properties to a stakeholder. Software engineering has some (though not enough) relevant evaluation techniques but lacks frameworks to compare, develop, and apply those techniques in a manner that respects h value varies by stakeholder. We present an adaptation of economists' value models that, given a design and a development method, predicts value to a client. We give examples supporting our approach. Even in its preliminary state, our approach helps to explain and characterize design evaluation techniques and shows sufficient promise to justify further development.

Duke Scholars

Author Ashish Arora Fuqua School of Business