Building a Fuzzy Knowledge System for the Quality of Post-Press Processes Execution

Abstract

Based on expert evaluation, post-print processing of book publications is represented by a function with the following arguments: notebook production, binding, assembling, cover material, book block processing, decoration, final processing, and project. The main partial quality indicators of post-print processes were identified: the quality of preparatory and binding processes, and the quality of cover-making processes. The quality of preparatory and binding processes is represented by such linguistic variables as the post-print processing project, notebook production, assembling, and binding. The quality of cover-making processes is represented by the linguistic variables: book block processing, cover material, decoration, and final processing. Term sets and universal sets of values were defined for each linguistic variable at the lowest level. A multi-level model of fuzzy logical inference for forming the integral quality indicator of post-print processes was developed, illustrating the research hierarchy regarding the development of a fuzzy knowledge system. Numerical values of the membership functions of linguistic variables were obtained for the established ranks of the respective terms at five division points.

A fuzzy knowledge system was developed regarding the quality of post-print processes based on fuzzy set theory. A fuzzy knowledge base was constructed using expressions like "if — and — then," "if — then — otherwise," "if — or — then — otherwise," a knowledge matrix, and fuzzy logical equations for the highest level "quality of post-print processes" for the terms "low," "medium," and "high," as well as for partial quality indicators. The fuzzy set was transformed into specific numerical values. The fuzzy knowledge system enables the calculation of the integral forecast quality indicator of post-print processes.