JSAE Paper No 20005335 Presentation No. 159 Speaker’s name R. Happee page 1 159 Analysis of rear end impact using mathematical human modelling R. Happee*, R. Meijer*, K. Ono**, M.J. van der Horst*, K. Yamazaki** *TNO Automotive Crash Safety Centre, Delft, The Netherlands **Japan Automobile Research Institute, Tsukuba, Japan At TNO an omni-directional mathematical human body model has been developed. Until now this human model has been validated for frontal and lateral loading using response data of volunteer and post mortem human subject (PMHS) sled tests. For rearward loading it has been validated for high speed impact (10 to 15g) using PMHS responses. The objective of this study was to simulate JARI volunteer low speed rear impact sled tests using the MADYMO 50 th percentile male human model. From the simulation results of the JARI volunteer sled tests together with simulation results of high speed rear impact PMHS sled tests, it can now be concluded that the human model is a valuable tool to predict head and neck responses to low and high speed rear impact. Keywords: Human model, Rear impact, Validation, Volunteers. INTRODUCTION In the current crash safety design and research mathematical modelling is widely used in addition to mechanical testing. Most occupant models used in crash simulations are based on crash dummies and thereby inherit apparent differences between dummies and the real human body. At TNO a mathematical human body model representing a 50 th percentile male has been developed. The human geometry was obtained from RAMSIS anthropometric data which provided a realistic and detailed surface description, in particular for seated automotive postures. The 50 th percentile male model from RAMSIS has been converted to MADYMO and extended to allow crash simulation (Happee et al. 1998). Many occupant models are only capable to simulate the response of car occupants for a single loading direction. The MADYMO human model has been developed aiming at omni-directional biofidelity, where the highest priority was given to the torso and the head-neck system. Until now this human model has been validated for frontal and lateral loading using response data of volunteer and post mortem human subject (PMHS) sled tests (Happee et al. 1998, 2000). For rearward loading it has been validated for high speed impact (10 to 15g) using PMHS responses (Horst et al. 2000, Bertholon 1999). The objective of this study is to simulate low speed rear impact volunteer sled tests performed by the Japan Automobile Research Institute (JARI) using the MADYMO 50 th percentile male human model. VOLUNTEER TESTS At JARI low speed rear impact sled tests have been performed with 9 adult male volunteers (Ono et al. 1999, Davidsson et al. 1999). Their average age was 26 years (SD 4 years). Their average height was 1.76 m (SD 0.3 m), and their average weight was 71 kg (SD 6 kg). A rigid seat (R8R) was mounted on the sled at an angle of 10 degrees with the horizontal, and the back of the seat was at 110 degrees with the horizontal. The feet were on a foot-plate at 45 degrees with the sled. No belts were used during the sled tests. In Figure 1 the sled test apparatus is shown. The maximum sled acceleration in the tests was about 3.6 g. The mean sled acceleration (in the sled direction) with standard deviations are shown in Figure 2. Each volunteer was asked to hold the handle and relax during the test. The test was repeated several times for each volunteer, so that they got used to it and were able to relax. The volunteers were fitted with film targets and accelerometers for measuring head and T1 kinematics and dynamics. The locations of the film targets and accelerometers were determined from x-ray and high speed film images of the instrumented volunteers. Figure 1: JARI Sled test apparatus