TABLE OF CONTENTS The subject of this section represents an intersection of the disciplines of software engineering and reliability engineering. Current information on the methodology and tools that software engineering has contributed can be found in the Institute of Electrical and Electronics Engineers (IEEE) Computer Society's Transactions on Software Engineering [12] and in publications by the Software Engineering Institute (affiliated with Carnegie Mellon University) [13]. We focus on those methods and tools that address the handling of REs that Section 7.2 identified as a major cause of failure in systems that had been extensively screened. An important contribution to avoiding these failures could come from a better statement of requirements, and that will be this section's first topic. We will then discuss improvements of test techniques and finish with a brief description of UML-based software development that holds promise in helping with both requirements and testing. Between requirements and tests is the entire area of software design and coding, which has been omitted here because it is well covered by software engineering.
Many of the failures previously discussed are frequently due to missing, vague, or incorrect requirements, particularly with regard to exception handling. System engineers and analysts are much more motivated to specify in great detail what the system should do under its normal conditions of service than what it should do (or not do) under rare conditions.
For example, consider the temperature monitoring of a high-pressure steam line. Three transducers are mounted on the line,...
