Abstract
The objective of this paper is to show through a real-world application why qualitative methods based on interval (or fuzzy) arithmetic with strong properties concerning soundness and completeness are adequate for modeling complex continuous processes. Firstly, we discuss modeling characteristics of such an important class of physical systems that includes chemical, nuclear, siderurgical and other industrial processes. On one hand, although it is almost always impossible to define numerical models for complex processes, it is usually possible to define boundaries (intervals) for the system parameters. On the other hand, some precision for the simulations is always required. We show that only interval (or fuzzy) based methods (and not pure numerical or qualitative methods) are adequate. Besides, for the effective use of such methods, soundness and completeness properties are of great importance. Secondly, in order to justify our claims, we present the successful application of QFSIM, a particular fuzzy-based qualitative method, to model a complex siderurgical process at CST Companhia Siderurgica de Tubarao, a Brazilian-Japanese company located in Vitoria-Brazil. Transactions on Information and Communications Technologies vol 1, © 1993 WIT Press, www.witpress.com, ISSN 1743-3517 280 Artificial Intelligence in Engineering
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
