ABSTRACT
Direct measurement of human performance is useful in many different ways, as discussed in the previous chapter. However, this approach may be infeasible or prohibitively expensive in some situations. This chapter describes three other approaches, which can be used to predict human performance. The use of Synthetic Data or Predetermined Time Prediction Systems involves an initial task analysis of job-related activities. Time requirements for the activity are then predicted by assigning values from tables containing expected times for elemental tasks. Standard Data Systems use historical data to predict the time needed for a family of jobs. This process involves developing a multivariate regression from the available data, and then using the equation to predict performance time. The third approach is to apply modeling techniques drawn from the field of cognitive engineering, which focus on the underlying cognitive processes of job-related activities. The latter approach has similarities to the first two, but requires significantly more effort on the part of the practitioner.
