A Framework for Model-Based Adaptive Training

Training System - Research Summary

The goal of this work is to develop a specification and realisation framework for using multiple explicit models of cognitive processes and qualitative models of a physical device, product or process for industrial training applications.

Few methodological approaches for building model-based training systems in industrial environments exist. The framework identified and presented here is made up of specification and realisation methods in terms of four core activities:

1. tried and tested analysis methods for a training problem specification (i.e., methods which do not consider a computer based realisation);
2. analysis methods for task decomposition, expertise classification and trainee characterisation (i.e., methods which consider a model-based computer based realisation);
3. design methods for intelligent training system realisation which separate what is to be taught (called subject expertise) from the way that it is to be taught (called training expertise); and
4. design methods for realisation of expert and trainer agents which consider multiple modelling dimensions at a conceptual level and a software implementation level.

Since all of these methods are developed within the same framework, they work together towards the ultimate goal of delivering Model-Based Adaptive Training as a new solution for industrial training situations. A specific capability within the framework, promoting adaptability, is in the flexibility of mapping problem spaces, domain specific training tasks and problem solving methods to a set of seven generic modelling dimensions [ITSIE 1992, D7]. A demonstrator system has been developed and the mapping of the elements that make up the framework is analysed for typical industrial training situations.