Some Models of Human Error for Man-Machine System Evaluation.
Abstract
A framework is presented for quantifying and modeling the occurance of human error events in a system. Decision tasks are classified according to the stochastic behavior of the relevant variables associated with the man-machine linkages. A general semi-Markov formulation is presented for describing transitions among error states. Closed form results for the time-between-errors are given through transform relationships for the case where interdecision times form a renewal process. An approximate time-between-error distribution is given that provides computational ease and compared well with some experimental data obtained from a laboratory human decision-making task. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- May 15, 1979
- Accession Number
- ADA072838
Entities
People
- Marlin U. Thomas
Organizations
- University of Wisconsin–Milwaukee