Home > Textile News > Expert system for textile and coloration industry

Expert system for textile and coloration industry

Published: 2019-05-21

‘Expert systems’ is a branch of artificial intelligence that makes extensive use of specialized knowledge to solve problems at the level of a human expert. The famous mathematical physicist Feigenbaum has defined an expert system as “an intelligent computer program that uses knowledge and inference procedures to solve problems that are difficult enough to require significant human expertise for their solutions.”

Expert system for textile industry

An expert is a person who has expertise in a certain area; the expert has the knowledge or special skills that are not known or available to most people. An expert can solve problems that either most people cannot solve or may solve but less efficiently.

The knowledge base stored in an expert system can either be the expertise of an expert or knowledge that is available from books, journals and knowledgeable persons. The expert system can be characterized in many ways such as knowledge-based expert system, fuzzy expert system, etc.

Expert system for coloration and textiles industry

Expert systems in coloration and textiles Expert system technology has been explored in different fields of textiles, including color grading of cotton, spinning, drafting, fabric formation, textile finishing, garment manufacture and technical textiles. The level of achievements in developing expert systems in computer-aided production management of spinning machinery is assessed elsewhere.

A knowledge-based system has been developed for spinning management which can be used either to predict the characteristics of the yarn according to those of the raw material or to select the raw material to produce yarn with specific characteristics.

An expert system has been developed for fault diagnosis of a filament yarn spinning machine and another, Corosult, for selecting the best way to operate Autocoro.

Sandoz AG developed Wooly, an expert system for commercial textile dyeing that could predict fastness performance to a wide range of standard tests and recommend details such as the processing route, dyeing methods, suitable dyes, and is capable of being interfaced with a color matching system.

Texperto, developed by Clariant, was designed for textile finishing and its performance and flexibility is described by Frei and Poppenwimmer.

Bafarex by BASF for pad steam dyeing and polyester exhaust dyeing.

SmartMatch for color matching by Datacolor International.

Calopoca for color matching by Ciba, for dyeing recipe determination and optimizing lab—to—bulk reproducibility, and Datawin which is a dyeing control system.

Expert system technology has also been explored in finishing, including a selection of fluorescent whiteners, waterproof breathable clothing and for oil and water repellency.

An expert system for total quality management has been developed for use by manufacturers before the accreditation process to save time and reduces consultancy fees. The prediction of fabric sewability and the design of geo—synthetic in relation to unpaved road design [134,135], are amongst the areas where expert systems have been explored.

In weaving, expert systems have been employed to form the basis of a management procedure for weaving problems, the objective being to find the reason for each incident and to keep in memory all the information concerned [136]. Some other expert systems have been developed for identifying fabric defects.

Modex [modular expert system] is a knitting expert system which gives the user the possibility to automatically create an executive program for the flat knitting machines from Stoll, Universal, Shima Seiki, etc., as well as machines from Scheller.

Tess, an expert system for diagnosing defects in woven textiles, has been developed by Swiss Federal Laboratories for Materials Testing and Research. The system which took a period of five years for its development was originally based on the idea to retain the knowledge of a damage expert employed at EMPA who was due to retire.

Keeping in view the interest of the companies in the initial prototype, the implemented knowledge—base of the system was extended to 12 partners including, spinning and weaving mills, textile finishing companies, manufacturers of colorants and auxiliaries, manufacturers of textile machinery and test equipment.

Basic concept of an expert system

Basic concept of an expert system where the user supplies facts or other information to the expert system and receives expert advice or expertise in response. Internally the expert system consists of two main components. The knowledge base contains the knowledge with which the inference engine draws conclusions. These conclusions are the expert system’s responses to the user’s queries for expertise.

Basic concept of an expert system

Figure 1: Basic concept of an expert system.

Development of an expert system

The processes of building an expert system is called knowledge engineering and is carried out by a knowledge engineer. Knowledge engineering refers to the acquisition of knowledge from a human expert or another source and its coding in the expert system.

The knowledge engineer first establishes a dialogue with a human expert in order to elicit the expert’s knowledge. The knowledge is then coded explicitly in the knowledge base. The human expert then critically evaluates the engineered expert system and provides an assessment to the knowledge engineer. This process iterates until the system’s performance judged by the expert is considered to be satisfactory.

Development of Expert System

Figure 2: Development of Expert System.

Basic components of an Expert System

  • User interface: the mechanism by which the user and the expert system communicate.
  • Explanation facility: explains the reasoning of the system to the user.
  • Working memory: a global database of facts used by the rules.
  • Inference engine: makes inferences by deciding which rules are satisfied by facts or objects, priorities the satisfied rules, and executes the rule with the highest priority.


  • Expertise could be available on any suitable computer hardware;
  • The cost of providing expertise per user is greatly reduced;
  • Unlike human experts, who may retire, quit or die, the expert system’s knowledge will last indefinitely.
  • Expert systems increase confidence that the correct decision was made by providing a second opinion to a human expert or break a tie in case of a disagreement by multiple human experts.
  • Answers to queries can be obtained in a very short period of time, leading to a fast response process.


A practical limitation of many expert systems today is the lack of causal knowledge. That is, the expert system does not really have an understanding of underlying causes and effects in a system. It is much easier to program expert systems with shallow knowledge based on empirical and heuristic knowledge than with deep knowledge based on basic structures, functions and behaviors of objects.

Another problem with expert systems today is that their expertise is limited to a knowledge domain that the systems contain. Typical expert systems cannot generalize their knowledge by using an analogy to reason about new situations the way people can.

In spite of their present limitations, expert systems have been successful in dealing with real-world problems that conventional programming methodologies have been unable to solve, especially those dealing with uncertain or incomplete information.

Expert systems building tools and shells

A very important development of modern expert system technology is the availability of different tools and shells, in order to ease the process of building expert systems. A tool is a programming language plus associated utility programs to facilitate the development, including debugging and delivery of application programs.

Expert systems building tools

A shell is a special purpose tool designed for certain types of applications in which the user must supply only the knowledge—base. This means that an expert system does not have to be built from scratch for each new application. A large number of such expert system tools are available at a very cheap price (some are even free of cost to download, but a really sophisticated tool might cost more.)


Artificial intelligence and expert systems have been successfully used in a number of industrial sectors. The inclusion of these systems has been shown to accompany added benefits, reduce overall costs and increase efficiency and output. Despite a number of examples given on the use of these systems in the coloration sector, their full potential has not been exploited.

Perhaps one of the reasons for the limited enthusiasm in utilizing these systems in the coloration sector is the seemingly complex nature of both domains of knowledge and the ‘know—how’ attitude in the domain. Competitiveness is the key factor in the survival of this industry. It is therefore obvious that efforts should be directed towards reducing costs and increasing output. This has been traditionally achieved by reducing labor cost and increasing automation.

However, the expertise of masters who have worked for a lifetime in the sector cannot be easily automated. This information is normally lost unless efforts are made to store the knowledge and expertise in a suitable system that allows easy retrieval and usage.

To that end, this has not been fully achieved. This paper has reviewed a number of approaches for the modeling and construction of artificial intelligence and expert systems. As it was stressed, the choice of each system will depend on a number of factors and it is useful to ask a series of questions to verify whether they would be useful.

It is our belief that most sectors of the coloration industry would benefit from expert systems. These systems could be constructed by computer specialists in conjunction with experts in the domain to provide in—house expertise.

Alternatively, domain experts familiar with programming languages and available shells could construct their own expert systems. Obviously, this would take time and it is recommended that a partnership with artificial intelligence experts be sought. Troubleshooting diagnostic system initially for the coloration of cotton is under construction at the School of Textiles of Heriot-Watt University.

This system is based on a hybrid approach utilizing the effectiveness and similarity of fuzzy logic to real scenarios in conjunction with expert system technology. Most of the terms used in the color industry are vague and this information can be computed by fuzzy logic algorithms.

Although it is hoped that the industry will flourish and remain competitive, we believe that industrialists and experts must rise to the challenge and jointly reap the benefits of these systems to avoid witnessing a widening gap in the shortage of expertise in the sector in the near future.
















Source:Textile Today