Testability is a design characteristic which allows the status of product to be confidently determined in a timely manner. It addresses the extent that a product supports Fault Detection (FD), which is a process that discovers the existence of faults, and Fault Isolation (FI), which is a process that identifies one or more replaceable units where the faults (discovered during FD) may be located in a positive, favorable, and cost-effective manner.
Testability analysis is a method for evaluating the various qualitative and quantitative aspects of the Built-In Test (BIT)/Built-In Test Equipment (BITE) design. The various qualitative and quantitative aspects of BIT/BITE are evaluated, including FD probability, BIT coverage, FI resolution, and False Alarm (FA) probability. FA is an indication of a fault where no fault exists, e.g., operator error or BIT deficiency.
Why do we need testability analysis?
There are many beneficial reasons why testability analysis needs to be accomplished. The need for testability analysis may be required by product requirements (statement of work, performance specification, etc.), to increase FD and FI, and/or to reduce repair time for maintenance personnel.
A common requirement is for a product to be able to detect a certain percentage of faults during the different types of BIT. Another common requirement is to isolate faults to a certain number of Shop Replaceable Units (SRUs) that make up the product. An SRU is a modular component that is designed to be replaced at a back shop. Performing a testability analysis will satisfy the need for the product to meet the BIT and/or Self-Test requirements.
The testability analysis will also identify components not covered by BIT. The products design, BIT, or combination of the two can be enhanced to improve the self-test diagnostic capabilities to detect and report more faults, which creates a better product.
Additionally, the repair time for the maintenance personnel will be reduced when failures can be easily isolated to an SRU or subset of SRUs. The reduction in repair time results in cost savings for the product line.
What other specialty engineering analyses are used in testability analysis?
The inputs required for an effective and efficient testability analysis include data from Failure Mode, Effects, and Criticality Analysis (FMECA), types of BIT, and tests covered in each type of BIT. A Failure Mode and Effects Analysis (FMEA) is a procedure for analyzing each potential failure mode in a product to determine the local and end effects of a failure on the product. When the analysis is extended to classify each potential failure mode according to its severity and probability of occurrence, it is called a FMECA.
The FMECA provides failure modes, failure detection methods, and severities associated with a product. The FMECA also identifies when the effects of a failure can be detected using BIT. The alternative to utilizing a FMECA are to develop BIT tests based on knowledge of how the product works, and not whether a BIT test needs to be developed for a particular fault that is critical to product performance.
The types of BIT and their associated tests are important to determining whether a product meets the testability requirements. The types of BIT include:
Performing a Testability Analysis should be designed and conducted early in a program so that improvements or additions into the BIT hardware, tests, or procedures can be incorporated into the design. It is important that the product team including design engineers, test engineers, maintenance personnel, specialty engineering and other key stakeholders to provide support throughout the process. The testability analysis process will not only satisfy program requirements but also yield a highly maintainable product.
Article Authored by Antonio Scappaticci