ACCIDENT INVESTIGATION REPORTS and their supporting data constitute a major portion of the information we have on how humans make errors in real world situations. Unfortunately, the very nature of most accidents hampers our search for the causes of the errors. Laboratory research tells us a lot about human behavior, but if it gave us all the answers we wouldn't have accidents to investigate. Since accident investigations are not laboratory experiments, our methods of data collection and analysis are different, and our experimenters (investigators) have different skills and qualifications. At the National Transportation Safety Board (NTSB), we have gone through an evolution of thought on the philosophy of human performance investigation--in fact we are still evolving. The following are some issues we've been faced with, along the way. Some of the major issues are:
- Who should conduct human performance investigations?
- What facts should be collected? What should be ignored?
- How should human performance data be analyzed?
- Where does human factors research come into the equation?
How we approach these issues certainly affects what we find out about the contribution of human error to an accident, which in turn affects the management of human error, through the actions resulting from NTSB findings and recommendations. The purpose of this paper is not to answer these questions. Instead, I hope to rouse some thought and discussion.
File Size: 534K
Product Status: In Stock
Purchase more technical papers and save! With TechSelect,
you decide what SAE Technical Papers you need, when you need them, and how much you want to pay.
Learn more >
