Abstract
Scaled dummy used in the experiment on the estimation of passenger's behavior under a train collision is designed considering a similarity rule. Several tests are applied to evaluate dummy biofidelity according to ISO technicalreport ISO/TR9790. Overall biofidelity of the dummy is 4.62 points out of 10. And it is the third-highest rating "Fair Biofidelity" out of 5 ratings indicating the degree of biofidelity. Next, Finite element is applied conducted to improve thoraxbiofidelity dummy using optimization method, suchas breadth of rib, thickness of soft tissue and Elastic Modulus of soft tissue because response on thoraximpact test is particularly bad. Optimized dummy is developed and thoraximpact test is conducted again. The maximum value of impactor force of optimized dummy is 28.1% better than that of unoptimized one.
