Dependency Models
A model for Diagnostics, Prognostics and Testability

Areas of Applicability: (comparison chart)
 Diagnostics / Prognostics  System Engineering  Maintainability
 Testability  Reliability  Life Cycle Cost
 Design-for-Test  Safety  Other

What - Dependency Modeling as a means of solving Diagnostic and Testability problems began in the 1960's. The innovation was that by using cause-and-effect reasoning, a process could be devised to detect and isolate faults based on how they could be observed.

Why - Dependency Models can be used to solve diagnostics, testability and even prognostics problems. They also have some applicability to solving Design-for-Test, System Engineering, Reliability, Maintainability and Life Cycle Cost problems. This is because they can be used to directly answer questions about fault detection, isolation and other fault group statistics.

When - Dependency Models can be used at any phase in the design process, but because they represent

History - Dependency Models were first created for Diagnostics by Ralph De Paul Jr., founder of DSI International. Initially called Logic Modeling, the approach became the basis of all of today's model-based testability tools. Over the years, it was the source of much debate as to its capabilities, because it is such a low-level way of describing cause-and-effect relationships. It is primarily the lack of it being a high-level approach, as opposed to there being any specific deficiencies, that has lead to today's more modern high-level modeling approaches.

Related Links - Functional Dependency Models, Failure Mode Dependency Models, Hybrid Diagnostic Models