izvorni znanstveni rad
Advisory-based Product Configurator used in Chiller Configuration and Evaluation
Sažetak
This article presents the architecture and functional principle of an advisory-based product configurator. The program is used to create new and modify similar product configurations and is intended for users without in-depth expertise in technical fields. Due to its architecture, the configurator is designed to utilise the knowledge stored in the knowledge database. The selection is based on a combination of fuzzy inference rules and mapping techniques. Knowledge regarding the product is captured by using component-based matrices. The possibility of adding new knowledge and new rules is also provided. All this enables the configurator to link customer requirements and offers the possibility of selecting one or more products. In addition, it evaluates possible product variants based on the parameters specified by the user. The graphical representation of the results obtained includes the product technical documentation as well as a matrix representation of all the selected components and modules. The functional principle and evaluation of the product selection are demonstrated using a case study of a chiller for installation in an industrial plant. During the selection and evaluation of the chiller, better system performance is achieved that has a chiller with a water-based condenser.
Ključne riječi
product configurator architecture, fuzzy inference rule, matrix representation, CCM methodology, product selection