Abstract The Wright brothers' first flight took place in 1903, and for the next 60 years flight was purely a mechanical activity. This all changed in the early 1960s, when computers first began to find their way onto airplanes. The first onboard computers were hybrid electromechanical devices and made use of very specialized and limited instruction sets. There was a substantial lag between the appearance of digital technology in military aircraft and those flown in civilian applications. Although many people view aviation as an extremely high technology field, there is often a lag in technology adoption, particularly in the airborne segment of civil aviation. This lag is the result of conservatism due to safety concerns as well as a need to constrain both the weight and the cost of onboard systems. While slow to embrace the digital age, aviation‐related applications of software engineering are now at the forefront of this new science. Software applications are now found in all phased of aviation and include among the many, aircraft design, flight safety, air traffic control, maintenance, etc. Ever greater levels of automation onboard aircraft and in ATC systems require highly complex designs while demanding the utmost reliability and dependability of the software and the hosting hardware. Time and experience has shown that such reliability is the result of disciplined, well‐planned, thorough development processes. Current regulatory guidance for airborne systems containing software invokes DO‐178B/ED‐12B as the design assurance standard of choice to ensure safe software. Where software blends with hardware to produce firmware in the form of application‐specific integrated circuits and programmable logic devices, a new standard, DO‐254/ED‐80, has been introduced to extend design assurance to this area of software engineering for airborne systems. The software engineering approaches and techniques discussed here are a result of ongoing debate and consensus among the industry, regulators, and researchers. There is much engineering judgment in the techniques described, although there is no empirical basis or measurement of how best these techniques can eliminate safety‐related problems. Models have been developed to predict the latent fault density by correlating the maturity of the development process. Such models have not yet been applied and tested in the aviation industry. It should be noted that as the amount of software has steadily risen in commercial aircraft, the fatal accident rate has steadily fallen over the same time period. While a direct cause–effect relationship has yet to be established between the design assurance approaches discussed in this article and overall system safety, the trends provide significant confidence in our modern aerospace infrastructure.
5
