The first part of this paper outlined the Statistical Agent-based Model of Development and Evaluation (SAbMDE) and demonstrated the model's ability to estimate development cycle resource utilization. This second part of the paper explores the model's ability to compute development cycle information content and process risk. Risk managers focus mostly on outcome risk, i.e., the likelihood that a running system will behave in an undesirable manner. SAbMDE assumes that a subset of outcome risks are not inherent and immutable but are, instead, the result of defects and vulnerabilities introduced during the system's development process. The likelihood of defect and vulnerability introduction is a process risk. SAbMDE further assumes that measuring process risk is a prerequisite for minimizing defects and vulnerabilities and, therefore, outcome risk. The model implements the measurement with Shannon's information-probability relationship similar to its use in Axiomatic Design Theory (ADT). This paper details the SAbMDE's information and risk calculations and demonstrates those calculations with examples. The process risk calculation is consistent with and offers a mechanism for the ADT Information Axiom.
- Axiomatic Design Theory
- Decision Theory
- Design Theory and Methodology
- Information Theory
- Risk analysis