Software Fault Tree Analysis of an Automated Control System Device Written in Ada

Software Fault Tree Analysis (SFTA) is a technique used to analyze software for faults that could lead to hazardous conditions in systems which contain software components. Previous thesis works have developed three Ada-based, semi-automated software analysis tools, the Automated Code Translation Tool (ACm) an Ada statement template generator, the Fault Tree Editor (Fm) a graphical fault tree editor, and the Fault Isolator (Fl) an automated software fault tree isolator. These previous works did not apply their tools on a real system. Therefore, the question addressed by this thesis is 'Do these tools actually work on a real-world software control system?' This thesis developed and implemented a sample Software System Analysis Methodology (SSAM) using these semi- automated software tools. The research applied this methodology to a real-world distributed control system written in Ada. The Missile Engagement Simulation Arena's (MESA) control software was developed by the Naval Air Warfare Center, Weapons Division, China Lake, CA. The SSAM was used to show that the analysis of the Sphere-HWCI control module's 74,000 lines of code could be thoroughly analyzed in less than 100 man-hours. This practical, 740 lines-of-code per hour rate was a direct result of the incorporation of the semi-automated tools into the process.