A diagnostic expert system for the dyeing of protein fibres

Abstract

Coloration of protein fibres is characterised by many variables, each with a different effect on the final product. Because the process can be rather complicated, it is often difficult to achieve the right colour in the first dyeing attempt. Determining the root causes of a given problem tends to be even more challenging. While quality requirements in the textile industry have become increasingly rigorous, seasoned coloration experts have become rather scarce. This situation has exacerbated the need for the development and implementation of expert systems to augment available expertise in this domain. In addition, benefits associated with computer-based diagnostic systems have become increasingly evident over the past few decades, and the field remains an active area of research. Here we report the design and development of a diagnostic expert system for the dyeing of protein fibres. The system is designed to aid in the identification of root causes of problems with a view to enabling users to arrive quickly at remedial solutions. The performance of the system has been tested and evaluated by human experts and deemed to be highly satisfactory. This diagnostic system can be used to teach students, may be utilised by novice colourists as a problem-solving tool, and may be employed as a supplementary knowledge resource by seasoned dyers.