Abbssttrraacctt：In the past decade, researchers have identified that vehicle safety is an important factor when consumers are making decisions on vehicle purchasing。 However, different crash test protocols are used in different regions of the world to provide safety information to consumers。 From a homologation point of view, dissimilarity in these safety standards makes it very difficult for original equipment manufacturers (OEMs) to produce cars that can be sold around the globe。 Computer-aided design (CAD) and computer-aided engineering (CAE) tools have been widely used in modern industry to lower product development costs and shorten the design cycle。 Many design iterations can be evaluated on computers before a prototype is made for physical testing。 With ever-increasing computational power, numerical models are becoming indispensible tools in meeting the challenges prescribed by numerous, and sometimes conflicting, safety standards。 In this article, the following are discussed: the use of a hierarchical approach to validate CAE models, selection of the most appropriate design parameters for restraint systems through design of computer experiments (DOCE) methods, and the next wave of passive safety regulations。 Keey wwoorrdds: CAE；DOCE；finite element analysis；vehicle safety83010

A number of market research studies conducted by OEMs, insurance companies, and universities have identified vehicle safety as an important factor to consumers before making purchasing decisions [1]。 While other factors, such as price and style, also play an important role in finalizing the decision, noneare as precisely prescribed as crash test standards。 A common theme in various regions of the world is the New Car Assessment Program (NCAP)。Countries or regions have NCAP program include, but not limited to, Australia (ANCAP), China (C-NCAP), the European Union (Euro-NCAP), Japan (JNCAP), Korea (KNCAP), Latin American countries (Latin- NCAP), and United States (NCAP)。 These programs use different crash dummies, impact speeds, impact directions, moving or fixed deformable barriers, and pole configurations for testing vehicular safety。 As a result, it is very difficult for OEMs to design cars that can be sold globally using a single design。 Until a Global NCAP is established, OEMs need to continue designing different cars for different regions of the world。

Even if only one region (country) is considered, meeting the large number of crash safety standards can be quite expensive。 In the United States, for example, more than 20 standards are prescribed in the 200 series (for crashworthiness) of Federal Motor Safety Standards (FMVSS)。 Some of these crashworthiness standards, such as FMVSS 208, have a large number of test configurations。 As a result, traditional methods of cobbling parts together to meet crash standards becomes very expensive because of the high cost associated with hand-producing prototypes exacerbated by the large number of test configurations。

The computer-aided design (CAD) discipline can betraced back tothe graphicalcommunication

system developed by Sutherland in 1963 [2]。 Aside from graphical capabilities CAD tools provide, computer-aided engineering (CAE) tools also allow for the analysis of vehicular stress/strain and the performance of components, sub-assemblies, and full systems。 CAE tools allow safety engineers to run through amultiple number ofdesign cyclesbefore a final, optimal design is achieved。 As computers become more powerful, very accurate CAE models can be developed for this purpose。 However, such powerful tools can easily be misused if not properly scientifically validated。

One of the most commonly used CAE analysis tools is the finite element (FE) method。 This method is particularly suitable to model structures of irregular geometry, multiple material compositions, and complex loading and boundary conditions。 Stresses, strains, accelerations, and deformation patterns can be used for direct correlation with the tested structure。 In automotive safety, this method is widely used in the evaluation of vehicular crashworthiness and occupant safety。 The objectives of this paper are to use some examples to highlight some applications of FE models towards the design of a safer vehicle。