I-Promise—intelligent protective material selection
Introduction
In manufacturing, design is the process of transforming the customer's requirements into a set of specifications, drawings, plans, and other descriptions, suitable for conducting production operations and/or other practical activities in creating the corresponding artifact. Oakland (1989) defines the following phases in the design process, regardless of the kind and nature of design:
- •
conceptual design;
- •
detailed design;
- •
process and equipment design;
- •
prototype manufacturing.
Conceptual design means requirements identification and development of the product's specification. In many cases, conceptual design involves the product's configuration design. Such design has the following features (Brown, 1998, Franke, 1998):
- •
the artifact being configured is assembled from instances of a fixed set of well defined component types;
- •
components interact with each other in predefined ways.
This paper describes an application of expert systems in conceptual design of complex configuration tasks (or just ‘configuration’, for short). The industry considered is manufacturing of protective materials, and the system is deployed in “Tipoplastika” Holding Co., Gornji Milanovac, Yugoslavia. Complex protective materials are used for wrapping and packing of various products. In practice, complex protective materials combine a number of basic materials (such as different kinds of paper, PE foils, PP foils, PET foils, AL foils, etc.), in order to achieve desired protective features (e.g. water-proofing, oxygen-proofing, aroma preservation, and the like). It is a classic configuration task, but is simultaneously extremely complex and involves a lot of heuristics. Hence, we used expert system technology for designing complex protective materials. Specifically, our approach to such a configuration has two steps. In the first step, we consider the possibility of configuring the protective material starting from already used combinations of basic materials. If that is not feasible, in the second step, we try to generate an entirely new combination. Only the first step is described in this paper. The second step is the subject of our ongoing research.
Section snippets
Problem domain
Solving configuration problems in conceptual design is important in many branches of manufacturing. There are several critical issues here:
- •
customers can represent their requirements in a number of ways, hence flexible problem-solving strategies are necessary;
- •
constraints as to what parts can go together in the assembly can be very complex;
- •
the kind of activities that must be performed in order to get to a valid configuration;
- •
the schedule of the activities.
Manufacturing of complex protective
Problem statement
The problem we have focused on was how to properly configure a complex protective material starting from a standard combination of basic materials. This problem has the following steps:
- •
specify the customer's requirements as accurately as possible, allowing the customer to most effectively articulate his business needs;
- •
select a standard solution that most closely corresponds to the customer's needs;
- •
consider different commercial and technological constraints in the context of the selected
Previous work
The authors are not aware of any previous attempt to automate conceptual configuration design of complex protective materials by means of intelligent systems techniques. However, in our work, we have used experience and advice of authors and configuration designers from other application domains.
Approaches to performing configuration tasks differ in the methodologies and techniques they use (Faltings and Freuder, 1998, Sabin and Weigel, 1998). When configuration tools use intelligent
Proposed solution
In conceptual design, the solution space and the level of abstraction gradually decrease (Soininen et al., 1998, Wielinga and Schreiber, 1997). In our case, the solution is based on:
- •
a standard combination of basic materials;
- •
comparative evaluation of general protective features of the complex material with respect to the customer's requirements;
- •
comparative evaluation of those protective features of the complex material that can be measured, with respect to the customer's requirements;
- •
comparative
Design details
The processes shown in Fig. 1 involve numerous and often incomplete input data, heuristic reasoning and selection from multiple alternatives, and also possibly conflicting sets of requirements. This creates a good ground for applying expert system technology. In fact, I-Promise is designed as a series of several expert systems, Fig. 2. The link to the enterprise's information system is also shown.
Conclusions
I-Promise approach essentially covers two categories of the configuration task. The first one is configuration starting from existing similar cases; the second one requires using another technique. The goal is to solve as many configuration problems as possible by using existing similar configurations from the case base. The obvious benefits are:
- •
reduction of potential technological difficulties that new solutions often bring up;
- •
cost minimization;
- •
decreased design time;
- •
open end for using another
Mirko Đurić is the Manager of the Department of Advanced Technologies and Programs, ‘Tipoplastika’ Holding Co., Gornji Milanovac, Serbia, Yugoslavia. Tipoplastika Holding Co. is the largest manufacturer of complex protective materials in South-Eastern Europe. He has received his degrees from The School of Electrical Engineering, University of Belgrade, Yugoslavia (BS, 1982; MS, 1991). He is currently pursuing his PhD thesis in the field of intelligent systems. His 20-years experience in
References (23)
- Branting, L. K. (1998). Integration case and models through approximate-model-adaptation. Multimodal reasoning, Papers...
Defining configuring
Artificial Intelligence for Engineering Design, Analysis and Manufacturing
(1998)- et al.
Application of artificial intelligence and expert systems in industrial design: A survey
INFO Science
(1998) - et al.
Configuration
IEEE Expert
(1998) - et al.
Configuring large systems using generative constraint satisfaction
IEEE Expert
(1998) Configuration research and commercial solutions
Artificial Intelligence for Engineering Design, Analysis and Manufacturing
(1998)- et al.
Proceedings of the Seventh IEEE Conference on Artificial Intelligence Applications
(1991) - et al.
Knowledge-based evaluation of the conceptual design development of injection molding parts
Engineering Applications of Artificial Intelligence
(1996) - et al.
Case-based reasoning in design
IEEE Expert
(1997) - Marling, C., Petot, G., & Sterling, L. (1998). A CBR/RBR hybrid system for designing nutrition menus. Multimodal...
An industrial-strength description logic-based configurator platform
IEEE Expert
Cited by (0)
Mirko Đurić is the Manager of the Department of Advanced Technologies and Programs, ‘Tipoplastika’ Holding Co., Gornji Milanovac, Serbia, Yugoslavia. Tipoplastika Holding Co. is the largest manufacturer of complex protective materials in South-Eastern Europe. He has received his degrees from The School of Electrical Engineering, University of Belgrade, Yugoslavia (BS, 1982; MS, 1991). He is currently pursuing his PhD thesis in the field of intelligent systems. His 20-years experience in research, design and development of information systems is a valuable background for developing his current interests in artificial intelligence techniques and their application to information technologies.
Vladan Devedžić is an associate professor of computer science at the Department of Information Systems, FON—School of Business Administration, University of Belgrade, Yugoslavia. He has received all of his degrees from The School of Electrical Engineering, University of Belgrade, Yugoslavia (BS, 1982; MS, 1988; PhD, 1993). His main research interests include software engineering, intelligent systems, knowledge representation, ontologies, intelligent reasoning, and applications artificial intelligence techniques to education and medicine. So far, he has authored and co-authored about 160 research papers published in international and national journals and conferences. His major long-term professional goal is a continuous effort to bring close together the ideas from the broad fields of intelligent systems and software engineering. He has developed several practical intelligent systems and tools, and actively participates as a consultant to several ongoing projects in industry.
- 1
Tel.: +381-32-717260; fax: +381-32-717274.