Abstract
M.Ing.
In this thesis aspects of reliability management in a multi-level support environment are researched. Complex systems are generally supported over a number of support levels due to the specialist nature and support infrastructure requirements of the individual subsystems. Such a support approach also ensures optimum availability of the system whilst the subsystems are still in the repair cycle. Once a new system is put into service, it is exposed to the actual operational environment and not the simulated environment that was used to qualify the system during its development. In the operational environment, the system is also exposed to the support infrastructure. These factors, as well as any latent design and production defects, impair the achieved operational reliability of such a system. False removals and premature failures after a repair action further degrade the actual operational reliability of the system. It is generally not possible to qualify the logistic support infrastructure fully before placing a new system into operational service. Support stabilisation should take place early on in the support phase of such a system to correct all latent defects and deficiencies of any of the logistic elements required to support the system. Any latent design and production process defects not eradicated from the system will also surface during the support stabilisation period. Support stabilisation will ensure a constant failure rate for the operational life of the system at the lowest life-cycle cost. The methodology used to achieve system reliability growth during the support phase is similar to reliability growth during the development phase. However, additional variables of the operational and support environment are now included in the reliability growth process. The process is also further compounded by the geographic separation of the different levels of support each generally with their own support management infrastructure. The proposed approach is: get total management commitment and close the management loop over the different levels of support. establish the root cause of every system failure implement a test, analyse and fix policy eliminate ineffective repair actions ensure that the system operational environment is within the system specification remove latent design defects from the system correct deficiencies in the logistic elements.