Some time soon the Federal Motor Carrier Safety Administration will go public with its analysis of how to account for fault in the crash data it uses to assess carrier safety performance.

Joe DeLorenzo, director of the agency’s Office of Enforcement and Compliance, did not discuss the timing or details of the analysis in a CSA presentation Tuesday in Arlington, Va., but did say the study is done and has been peer-reviewed.

The agency is preparing a report to Congress and a Federal Register notice, he said in remarks to the CSA Subcommittee of the Motor Carrier Safety Advisory Committee.

The crash accountability issue is a hot button for trucking interests that object to the agency including non-fault crashes in the CSA Safety Measurement System, which measures safety performance. The agency takes this approach because its ability to distinguish fault is limited and there is a statistical probability that some of the crashes will be the carrier’s fault.

Meanwhile, DeLorenzo said, the agency is working on improvements to CSA.

One is to include in the database the results of court rulings on carrier challenges of roadside inspection citations. DeLorenzo said the agency is reviewing the more than 100 comments it received on this proposal and will publish an implementation plan this spring.

He said the agency also is working on improvements to the website where the public has access to CSA safety data.

The big CSA issue remains the proposed rule on carrier Safety Fitness Determination, the process by which the agency will use roadside inspection data and investigations to decide if a carrier is fit to operate. That proposal is scheduled for publication in mid-September.

Farther down the road, the agency is considering additional substantive revisions in the CSA system.

Dave Madsen, the principal truck safety engineer at the Department of Transportation’s Volpe think tank, said regulators are looking at changing the current approach to ranking the severity of violations.

The agency now assigns each violation a severity weight from 1 to 10, with 1 being the least severe and 10 being the most.

But that approach implies a level of precision in the data that does not exist, Madsen said.

A possible solution is to simplify the measurement system into three categories: low, medium and high, with additional emphasis if the violation leads to an out-of-service order.

The agency also is considering changing the way it groups carriers by the number of inspections they have had. The system measures a carrier’s performance in relation to other carriers that have had, say, from 5 to 10 inspections. If a carrier has 11 inspections it is moved to different category with carriers that have had from 11 to 20 inspections.

Problem is, the carriers in the new group have different performance characteristics. This is why some carriers see their scores change dramatically when they have additional inspections, even if the inspections find nothing wrong.

A possible solution, Madsen said, is to move from the tiered approach to smaller segments with, say, a two-inspection differential rather than 5 or 10 inspections.

This would eliminate the problem of large jumps in a carrier’s safety percentile when moving from one segment to the next, Madsen said.

The idea was well received by members of the subcommittee, whose general reaction was: “Why wouldn’t you make this change?”

Madsen replied that the approach is a little more complex for the agency but doable.

It may make its way onto the list of recommendations for CSA improvements that the subcommittee is preparing Wednesday. These will go to the full advisory committee, which will consider them for submission to the agency.