Embedded Computer Systems

The addition of more and more features to the Navy's F/A-18 "Hornet" fighter aircraft presents a major challenge to software development. The F/A-18, because of its ability to accommodate new capabilities, has become the lead platform for new avionics, weapons, and other subsystems. Currently, 28 other programs depend on the F/A-18 as their host platform. As a result, the Navy must accommodate the requirements and schedules of these programs as well as changes to the F/A-18's own software baseline. Integrating these subsystems into the F/A-18 while modifying the baseline requires major software changes. The competing demands created by other programs have already affected the F/A-18's software. Earlier software updates released to the fleets had problems and fleet-desired enhancements that were not corrected and implemented mainly because development schedules and software needed to integrate new weapons, avionics, and other subsystems were given higher priority. Navy officials say that despite these challenges they are now able to handle any F/A-18 software development by deferring the correction of less serious problems. GAO is concerned, however, that, as more and more requirements are levied on the F/A-18 in the future, the risk of serious software problems will escalate, challenging the Navy's ability to effectively manage software development.

GAO found that: (1) F/A-18 software development becomes more challenging as more capabilities are added to the fighter; (2) reasons for software changes include correction of problems during testing or development, enhancements to existing software, and integration of new avionics, weapons, and subsystems; (3) F/A-18 readily accepts new capabilities and is host platform for 28 other subsystem programs; (4) modifications to software development receive high, medium, and low priority rankings; (5) significant software modifications are required to accommodate the requirements and schedules of other programs and the F/A-18 software baseline; (6) previous software upgrades and fleet enhancements contained problems and experienced delays due to new weapons integration, avionics, and other subsystems given higher priority; (7) deferring correction of less serious problems could increase the severity of existing problems, become hidden in new software, and interfere with new software and subsystems; (8) a lack of modularity creates difficulties in identifying and correcting problems; and (9) reasons deferring problems and enhancements include priority to meet development schedules, unrealistically large amounts of software changes planned, and the cost verses benefits of fixing some problems cannot be justified.