Thinking beyond the group size fetish: towards a new testability

This paper exposes some major deficiencies inherent within current methods for assessing design testability-critical shortcomings that not only might cause adherence to contracted testability requirements to be in conflict with long-term maintenance goals (such as life cycle cost and operational availability), but could also result in evaluations that fail to predict the actual diagnostic behavior of the system or device. In recent years, the need to accurately forecast diagnostic performance has become more essential as more development projects are contractually linked to the maintenance of the fielded product (as witnessed by the recent emergence of maintenance warrantees and the combined contracting of development and maintenance efforts). What is needed are testability procedures that can better serve long-term maintenance and support objectives, yet remain true to the discipline's original intent of providing diagnostics-based feedback in early phases of the development cycle. Hoping to foster a less conflicted testability practice, this paper proposes some alternatives to current quantitative methods of testability assessment that can more accurately predict diagnostic behavior and more consistently reflect the relationships between a system or device's diagnostic capability and its life cycle cost and operational availability.