Electronic Thesis/Dissertation


Real World Derived Simulation Methodology for the Evaluation of Fleet Crash Protection of New Vehicle Designs Open Access

Downloadable Content

Download PDF

At the present time, the crash safety of new and concept vehicle designs, i.e., with advanced materials or structure or powertrains, is primarily assessed through computer simulations of single-vehicle crash test protocols specified by existing regulations and consumer information programs. Although such protocols are representative of real-world crash configurations, the tests are typically impacts into fixed object and are performed at single speeds with a single size of dummy occupants. In the real world, vehicles are involved in crashes with fixed objects and with other vehicles in the fleet. These crashes occur at various speeds and involve occupants of many sizes and ages. To date, assessment of the real-world safety of vehicle designs has not been attempted through simulations because a systematic approach is not currently defined or developed. In this research, a novel methodology for Evaluating Fleet, i.e., self and partner, Protection (EFP) of new vehicle designs has been developed through a systems modeling approach driven by structural and occupant modeling, and real-world crash and full-scale test data. The fleet societal injury risk in EFP is defined as the total injury risk of occupants in both the target vehicle and partner vehicles aggregated over a range of impact speeds, occupant sizes, and crash configurations, and weighted by relative frequency of the specific crash incident in the real world. The self-protection provided by a target vehicle is derived as the aggregate injury risk for its occupants in both single- and two-vehicle crashes. Partner protection is derived as the aggregate injury risk for the occupants of the vehicles in the fleet against which the target vehicle collides. The integral feature of EFP is that the methodology is based on real-world crash configurations, severity exposures and occupant injury risks in each crash incident, and is also based on physically realistic vehicle structural, occupant, and restraint system models. The main hypothesis of EFP is as follows: Given that the approach is grounded in the physical world with sufficient detail (i.e., real-world derived with sufficient granularity), EFP will serve as a useful method to assess the overall safety of vehicle designs in the fleet and for directing future vehicle safety research efforts.As proof-of-concept and initial application, EFP was implemented to assess real-world fleet societal risk in frontal crashes. The modeled frontal crash configurations and EFP weighting factors were derived from an innovative crash taxonomy based on real-world structural engagement from the National Automotive Sampling System Crashworthiness Data System (NASS CDS). The EFP crash configuration weighting factors for two-vehicle crash simulations were modulated by real-world crash exposure by vehicle class. The weighting factors for the impact speeds of the simulated crash incidents were based on real-world distributions for the target vehicle class. Simulation data to drive the methodology were obtained from finite element vehicle structural models with sufficient physical detail that allow implementation of new designs such as lightweight materials, new powertrains, and new structural architectures. Occupant responses were based on three-dimensional articulated rigid body models of the occupant and the passenger compartment. Both the structural and the occupant models are subjected to validation and robustness checks for the modeled crash configurations. Occupant injury potentials were derived from occupant responses using state-of-the-art biomechanical injury risk probability functions.EFP was applied to compute the change in driver societal injury risk between baseline and concept light-weighted vehicle design variants. EFP was also applied to obtain insights of the safety interactions in frontal crashes in an assumed light-weighted fleet as compared with a baseline fleet, both consisting of two vehicle segments. Overall, there is a net decrease in safety on a fleet level and by vehicle segments modeled in the light-weighted fleets. While the proof-of-concept and initial implementation of EFP was to drivers in frontal crashes, the EFP methodology can be extended to all crash modes and occupants.

Author Language Keyword Date created Type of Work Rights statement GW Unit Degree Advisor Committee Member(s) Persistent URL