Customers demand reliable products and such demands are reflected in the product specification and requirements documents. Indeed, contractual obligations are being introduced to ensure that the supplier meets the cost of unreliability. Thus, equipment suppliers must design products to perform reliably for the required product life. Moreover suppliers must provide sufficient evidence that their product will meet reliability requirements. This evidence is termed the reliability case. The main instigator for the use of the reliability case is the British Ministry of Defence (MoD) under the guise of Defence Standard 00-42 Part 3 but the concept is also becoming accepted in the commercial field as one of the key criteria for selecting products.

This paper describes the work carried out under a collaborative project funded by the UK Department of Trade and Industry (CARAD) entitled reliability enhancement methodology and modelling (REMM). More specifically it will concentrate on demonstrating the methodology in the form of providing the necessary guidance for building a reliability case. It will also discuss the challenges associated with design for reliability and demonstrate how the REMM tool will encourage designers to consider what might go wrong.

The REMM tool encompasses a reliability assessment model and this paper will describe and discuss the advantages of this Bayesian model with particular reference to decision making aspects during the design stage of a project.

High altitude span loader aircraft experience significant wing bending mainly due to turbulence of spanwise scale comparable to the wing. These moments are often the primary wingspar design loads. Spanloaders with a few concentrated loads on the wing are called semi-span loaders. Typical of this class is a family of unmanned solar-powered vehicles developed by AeroVironment of California, which have spans up to 250ft and flight speeds of about 30fts-’ equivalent. An aeroelastic model of the wing moments induced by turbulence varying in both spanwise and flightwise directions has been developed. The model treats flexural displacements and torsional deflection, runs rapidly in Excel® format, and can be used to study many different design and gust cases, thus it is a powerful tool for multi-disciplinary optimisation. The procedure was exercised for a generic semi-spanloader. Calculations were made for the effects of turbulent scale, of altitude, and of centrebody mass, all of which are shown to be small. Stiffness sensitivity calculations show that, although the composite wing is very flexible, the turbulence-induced moment is not greatly attenuated by flexibility. Effects of flexural deformation dominate the bending moments and torsional effects are very small. Seven different cases of the type used to define gust loads were examined. These indicated that gust models that do not contain spanwise variation seriously underestimate wing bending.

This paper sets out the case to extend the role of flight simulation in general aviation. It outlines the benefits that flight simulation has brought to airline safety but observes that the majority of pilots worldwide receive no simulator-based training or checking. The paper reviews the potential benefits of synthetic training and provides an analysis of accidents in general aviation to show that a significant reduction in fatal accidents is achievable by instituting recurrent pilot checks in a flight simulator, similar to the simulator checks undertaken by airline pilots. The paper summarises the advances in simulator technology which have occurred in recent years and outlines a scheme to fund the introduction of flight training devices in general aviation, based on compulsory recurrent checks for PPL and CPL pilots. The paper surmises that flight simulation offers a practical means to reduce the accident rate in general aviation and concludes by outlining a financial case to fund the introduction of mandatory simulation-based recurrent checks in general aviation.