Improvement of accident simulation model and improvement of narrow object accident reconstruction.

The basic objective was to assess the accuracy of the program, Simulation Model of Automobile Collisions (SMAC). This was done by staging three collisions and comparing the SMAC accident reconstructions to the empirical collision data. A second objective was the modification of the SMAC program to improve its accuracy for a narrow rigid object collision. Other objectives included (1) the reconstruction of 15 accidents involving air cushion restraint systems and (2) the incorporation of several modifications to the SMAC program which will increase the utility and applicability of the simulation. The three staged collisions included a 1970 Chevelle impacting a rigid pole, two 1973 Chevelles impacting in a front-to-side T-collision and 2 1973 Chevelles impacting in a front-to-side oblique collision. Each collision was reconstructed using both preimpact and postimpact data to determine the accuracy of the SMAC program. The first staged collision involved a rigid narrow object which required an extensive modification to the original SMAC program. The modified SMAC simulation gave results which correlated exceptionally well with the first collision. The last two collisions were reconstructed using the original SMAC program. Most of the comparison parameters (positions, velocities, accelerations, and damages) from the SMAC reconstruction compared favorably with the results from the staged collision: however, a few did not.