What is CRASH OLDMISS, the Missing vehicle Algorithm?

[MSI]

Since the late 70’s, NHTSA, as part of their NASS statistical studies of collisions throughout the nation, has done research on trying to reconstruct a collision when there is a missing vehicle. The most recent publication on the topic by NHTSA was called
The Missing vehicle Algorithm (OLDMISS]) Reformulation,
SAE paper 91-0121 by Prasad .

• NHTSA was trying to quantify the approximate range of impact speed changes for vehicles in injury producing collision for their NASS study (National Accident Statistical Study) and in some incidents they did not have access to one of the vehicles so they came up with an approximation technique to permit some form of reconstruction. The technique basically uses the detailed information on one vehicle, and knowing the type of other vehicle it allows you to approximate the crush on the other vehicle and therefore approximate the impact speed change.

From the study:

• “The results of the Calspan investigation of the OLDMISS model are confirmed in that it is found to be moderately accurate for front-to-front collisions, less accurate for front-to-rear collisions and severely inaccurate for front-to-side collisions”
• “Vehicle specific crush characteristics were found to provide some improved accuracy in all 3 collision modes”
• “As a result of its narrow application and inaccuracies, NHTSA does not sell o provide the OLDMISS program in any form to the public”

In a 1997 SAE paper by Grimes, et al entitled “Developing a Crush Profile Estimate by Balancing Impact Forces”, SAE paper 97-0942 the authors made no reference to the prior NHTSA study research but generally were performing the same task.
The 1997 paper included the following:

• “This method requires a detailed damage profile on at least one of the vehicles, accurate stiffness data for both vehicles, and a good understanding of the methods and underlying assumptions”
• “Several restrictions to this method:
  • Tire forces must be negligible
  • Accurate crush stiffness data must be available for the selected vehicles
  • Accurate PDOF values for both vehicles must be available
  • Accurate crush data for the vehicles with the known crush profile is required
  • There must be some way to estimate the damage center offset on the vehicle with the unknown crush profile.
• “Additional work needs to be done using a large database of cases before any statistical confidence levels can be established. A detailed sensitivity study on all the variables also needs to be completed”