A new reliability prediction method under digital development environment

During the product development lifecycle, the traditional approach to reliability specification has been based on often unrealistic reliability prediction by using commercial reliability models, such as MIL-HDBK-217 in the specific case of electronics equipment. Often, the prediction methodology used assumes an exponential failure rate, meaning that random failures and faults are inevitable. Therefore, without the recognition that many failures can be prevented by attention to basic design details. In today's context of global competition, manufacturers are facing greater challenges than ever before. Customers demand more complex and reliable products to be developed with shorter lead times and more cost effectiveness; Environmental and regulatory constraints as well as market expectations demand more efficient product behavior. In order to provide the customer with equipment that works when needed and continues operating for a defined period of time, manufacturers need to better understand materials and process conditions, and their effects on product reliability. However, traditional reliability prediction method, which is based on the fault logics of product composition units, has inherent limitations, e.g., it is not relevant to product design parameters. Finding new reliability prediction method to address the challenges that we face during the product development lifecycle is needed. In this paper, first, using the geometric data, material data and structure data provided by the geometric digital prototype of the product, we analyses the component reliability of the product by structure simulation. Then taking the results of structure simulation as inputs to the functional digital prototype and digital performance prototype, we analyze the product reliability by functional simulation. It is shown that this new reliability prediction method can accurately predict the reliability in the development phase.