The George Boole Foundation
The Decision Analysis Initiative



Learning, the everyday non-ending process...


Technique

What an individual learns is a technique or a way of applying technology or a tool in accomplishing some end. Technique occupies a space between a person's innate abilities and technology which can be divided into hardware and software. Indeed, the innate abilities of humans can also be divided into intellectual and learning capacity (software) was well as physical capabilities such as dexterity, good sight (hardware) and how the individual integrates these in general coordination in applying a technique.

The measurement of the process of learning is essentially the recording of the evolution or progress in the skill with which an individual applies a technique and even adapts a technique so as to improve performance both in terms of speed of accomplishment, quality of output and reduced wastage. Measures of learning include the use of less resources to accomplish output, less wastage, fewer rejected products, falling unit costs and production of each unit within a shorter time.
The Learning Curve

The learning curve is a measurable quantitative relationship between the number of arial that person has carried out a task and the competence of that individual in accomplishing the task. Thus it reflects the degree to which people learn through repetition. For any enterprise there tends to be a fixed quantitative gain from learning associated with each historic doubling of output.

Individuals & groups

Whereas this phenomenon can show the relationship between repetition and skill for an individual it is generally applied to industrial processes where several people are involved in different sub-processes applying different technologies with different appropriate techniques and varying skills in their application.

Factor-intensity and learning

In industrial processes, the more a process is automated (capital-intensive - see Factor Intensity) the less the learning curve effect.

Decision analysis considerations

There are two common applications of the learning curve:

  • analysis of production performance and projection of future costs projections
  • identification of performance enhancement opportunities achievable through investment in learning or automation
Simulation

Several learning curve algorithms have been developed for projecting future learning curve effects based upon measured historic data. Their common objective has been to measure and then simulate the historic doubling of production effect so that future cost performance can be estimated.

Performance enhancement opportunities

One of the most challenging issues facing organizational management is not only to determine the learning effect and project its impact into the future, based upon historic performance, but is also to analyse the source of performance gains within the process of learning so as to answer the questions:

  • where are there learning opportunities to improve performance?
  • what aspects of learning can and should be automated?
This is an important aspect of decision analysis applied to whole enterprise strategic planning (see Enterprise models & strategies) as a way to identify opportunities to secure gains in productivity, process yield, low reject levels and reduced unit costs.