Uncertainty in Software Development

``The Uncertainty Principle in Software Engineering,'' by H. Ziv and D.J. Richardson, May 1996

Abstract

In this paper, we define the Uncertainty Principle in Software Engineering (UPSE) and identify key domains where the principle is evident and applicable. We describe several examples of uncertainty in software engineering. We suggest that existing methods in artificial intelligence may be used to model and manage uncertainty in practical software development situations. We discuss how one such method, Bayesian Belief Networks, may be applied to scenarios of software testing and understanding. Finally, we discuss the implications and potential benefits of modeling uncertainty to software-process and software-product modeling.

from Debra J. Richardson
djr@ics.uci.edu
Department of Information and Computer Science, 
University of California, Irvine CA 92717-3425