Expert System for the identification of Plastic Materials

Mateo Lezcano Brito

Universidad Cooperativa de Colombia

Minerva Lezcano Brito

Universidad Cooperativa de Colombia

Lydia Ríos Rodríguez

Universidad Cooperativa de Colombia

Introduction: This research article is an update of the project “Expert System for the selection of plastic materials for the design and preparation of new products”, developed in 1990 by the Central University “Marta Abreu” of Las Villas. The objective of the current research, conducted in the year 2015 by the Cooperative University of Colombia, is to update the expert system developed in 1990.

Methodology: The Common kads methodology was used for the development of the expert system, which allows the structuration of knowledge by creating a distinction between different domains of application, task and inference, thus enabling the reuse of components. During the inference process, a search direction directed by objectives was used, due to the characteristics of knowledge contained in the bases, in which the inference process is initiated without preliminary information. 

Results: The newer version of the system improves the inferences, offers a more user-friendly interface and extends knowledge.

Discussion: The addition of knowledge bases, adapted to new situations, has an effect in the elicitation of better and more substantiating research answers, and the implementation of the new system over ucShell 3.0 provides a remarkable improvement in the interface, including the possibility of offering images and video during the inference process and admitting knowledge bases of any size while facilitating their exchange.

Keywords: reverse engineering, artificial intelligence, plastics, expert system
Metrics Loading ...