Programmed Safety through Programmed Learning

Abstract

A study was conducted to determine the relative effectiveness of various safety training techniques on industrial accident occurrence. A punch press simulator was designed and equipped with special electronic instrumentation to record the occurrence of simulated accidents as well as to indicate the production of substandard simulated parts. Four training techniques were used to impart instructions to four groups of 25 experimental subjects each. The training methods included: an audio-visual presentation (slides and audio tapes); direct, face-to-face oral instruction; no instruction (a control group); and a video tape presentation. The study proved that industrial simulation is a feasible means of determining the effectiveness of industrial training instructions. In terms of accident performance, the classical, face-to-face training method proved to be the most effective. There was a significant difference in accident occurrences relative to the four training procedures. There were no significant differences in the elapsed times to first accident, the total time required to perform the task, or the quality of parts associated with the various training methods. By constructing learning curves it was determined that the accident occurrence data conformed to a curve generally associated with learning while the production of good quality parts remained relatively constant.