The Need:

As our technology becomes increasingly complex, so does the quantity and variation of the Embedded Digital Device [EDD] components that are required within the project. Because not every device is made perfectly to specifications due to a propagation of random error, it is important to have a diverse, efficient testing process to avoid a facility's vulnerability to Common Caused Failure [CCF] (especially in the field of Nuclear Power).

The Technology:

A team of Engineers at The Ohio State University have researched and proposed a model to extend code-level mutation testing (Testing the functionality of an EDD relative to its Software Requirements Specification [SRS]) to the requirements and design level (Testing the functionality of an EDD relative to its Software Design Description [SDD]) by using a High level Extended Finite-State Machine figure [HLEFSM]. They then developed an Automated Mutation Testing Tool [AMuTT] that develops mutants, or changes, the logical gates, operators and operands in the software, and then tries to differentiate, or kill, the mutants by comparing the HLEFSM of a mutated EDD to the technical specifications of an ideal EDD, modelled by a HLEFSM of the SRS/SDD. A mutation score can then be calculated by taking the proportion of killed mutants to the total number of mutants generated.

Commercial Applications:

• Reduces 75% of the current cost of mutation testing while keeping a mutation score (effectiveness) above 99%.
• Can be marketed outside of nuclear engineering toward any organization that uses EDD's and is prone to CCF.

Benefits/Advantages:

• Develops efficient test suites to validate the SRS and SDD of an EDD.
• Automates the mutation test case process.
• Creates an executable model of the SRS and SDD.
• Only uses 5 of the possible 29 mutation operators.