The Monte Carlo method for non-detected failure analysis on dynamic systems testing

This paper presents the application of the Monte Carlo method for non-detected failure analysis on the mobile seat platform across the dynamic system testing on the commercial vehicle. This method has been applied in automotive embedded software of OEM components to reduce project development time, provide standardization and to predict software errors and look towards quality improvement. With growing customization needs and high number of components and special functions, the benefits of Model Based Design use tend to be explored at the highest level. A heavy vehicle (bus) OEM application of a Mobile Seat Platform (MSP) electronic control unit is explained in detail, with focus on Model Based Design method, and quality opportunities normally applied in automotive industry. An alternative loop simulation technique is proposed (Monte Carlo), as an extra step in the validation process to be carried out within the ISO 26262 V-model standard product development procedure. The method is useful to predict failure margins in mathematically modeled software blocks, due to parametric manufacturing variation mechanisms.