The Impact of a Learn-Forget-Learn (LFL) Curve and Learning Curves on A System Effectiveness Model.

摘要

In only the last 25 years has operator reliability been incorporated with hardware reliability to obtain a value for the effectiveness of a man-machine system. Very seldom, and in most cases, never, do any of the human reliability models address the effect that operator learning has on human reliability and the subsequent impact that operator reliability has on system effectiveness. This research studied the sensitivity of a system effectiveness model to changes in operator learning levels. Learning data which was expressed in terms of performance versus time, was obtained from a paper which analyzed the performance of an actual manufacturing task. This data was utilized to develop three different curves - a log pseudo-learning curve, a cubic pseudo-learning curve, and a Learn- Forget-Learn (LFL) curve. Each curve expressed operator performance as a function of time. The expressions for each of the three curves were then utilized in conjunction with a system effectiveness simulation model to formulate values for system effectiveness. The various values of system effectiveness obtained from the simulation demonstrated that the model was sensitive to changing levels of operator performance. This research is unique because this is the first time that operator learning curves have been utilized in conjunction with a simulation of system effectiveness. (Author)