A learning curve prediction model for operators performing industrial bench assembly operations

Abstract

Learning curves of operators performing industrial operations vary according to the characteristics of the operator and the characteristics of the operation being performed.In this paper a method is developed to enable prediction of a specific learning curve of the function Y = c+ a( 1-exp -b(t -L)) for a. specific operator performing a specific operation. This specific curve is based on certain measurements of the operator and certain measurements of the operation. IntroductionLearning curves can be characterized as graphical descriptions of the level of achievement (i.e. performance) attained during the learning period. Learning curves for industrial operations are generally of two types. The first type refers to the learning by an individual and is sometimes referred to as , acquisition of skill'. The second type refers to the learning by whole systems of men and machines and is usually referred to as the manufacturing progress function (Glover] 966).A number of authors have set forth various mathematical functions to describe learning curves. Most of these have been summarized by .Most learning curve research has focused on curves which have trials or repetitions or cycles on the X axis and time required to perform the cycle on the Y axis. The curve then indicates the time required for each cycle and indicates that this time decreases as the number of cycles increases.It would appear, however, that a more useful form of the learning curve for representing an industrial operator's output is one which has time Oll the X axis and output (in cycles per unit time) on the Yaxis. Specifically,' days' might be the units on the X axis and production output in ' units per day' on the Y axis. This is especially true where a company measures an operator's output on a time (e.g. daily) basis.Learning curves for industrial operators performing assembly tasks appear to be influenced primarily by the oharacteristios of the specific operator and the specific operation being performed.In this paper a means of predicting, in advance, the learning curve of a specific operator performing a specifie assembly operation is developed using certain' measurements' of the operator and' measurements' of the operation.

The Effects of Task Complexity and Experience on Learning and Forgetting: A Field Study

“…Bishu and Chen further concluded that greater performance improvements resulted from tasks performed in the order of difficult to easy than from those performed in the order of easy to difficult. Bohlen and Barany (1976) studied the direct effects of task complexity on learning parameters and concluded that task complexity affects The Effects of Task Complexity and Experience on Learning and Forgetting: A Field Study David A. Nembhard, University of Wisconsin-Madison, Madison, Wisconsin This paper examines the effects of task complexity and experience on parameters of individual learning and forgetting. Three attributes of task complexity and experience are addressed: the method, machine, and material employed.…”

Modeling and Analysis of Operator Effects on Process Quality and Throughput in Mixed Model Assembly Systems

“…A;^(l -e-^'") (9) where vector p=[pc PD PLY and ß=[ßc ßö ßiY. Figure 2 shows the effect trend of each of the three factors (pc, Po, and pi) on the quality conforming rate when the other two factors are fixed.…”

“…Traditionally, an operator's experience for performing a particular task refers to that operator's autonomous learning ability due to the repetitive execution of that task [9], While literature on learning curves abounds (e.g., Refs. [10,11]), there is limited research that addresses the effect of experience on quality conforming performance.…”

Progress functions: a general model

“…Since pre-production preparation is prior experience, particularly if the construction of prototypes is part of the prior effort, the novelty hypothesis becomes further strengthened. Goldman and Hart (1965), Bohlen and Barany (1976) and Kvalseth (1978) have all established that task complexity influences the characteristics of the learning curve. In particular, the study by Kvalseth performed in a laboratory situation permitted an independent test of the effect of task complexity (entropy) on the parameters of the learning curve.…”