Ontology-based data integration and decision support for product e-Design

Abstract

Currently, computer-based support tools are widely used to facilitate the design process and have the potential to reduce design time, decrease product cost and enhance product quality. Although there are promising information systems to manage product lifecycle and product-related data, including product data management (PDM) and product lifecycle management (PLM), significant limitations still exist, where information required to make decisions may not be available, may be lacking consistency, and may not be expressed in a general way for sharing between systems. Moreover, there remains little support for decision making that considers multiple complex technical and economical criteria, relations, and objectives in product design. To address these problems, this paper presents a framework for an ontology-based data integration and decision support environment for e-Design. The framework can guide designers in the design process, can make recommendations, and can provide decision support for parameter adjustments.

Introduction

Product design can greatly influence cost, quality and time to market of a product. Currently, computer-based support tools are widely used to facilitate the design process and have the potential to reduce design time, decrease product cost and enhance product quality. Design for Manufacturability (DFM) is developed with the expectation that it will address the time to market, quality and cost issue, and a Design for Assembly (DFA) system considers the ease of handling and assembling of component parts and the number of parts used in the product [1]. All DFM systems share a common goal, which is to minimize the total product lifecycle costs through more systematic and efficient decision making, when considering the design in light of manufacturing [2].

The data produced by existing information systems, such as CAD and DFM systems, are already in electronic formats. But all the information required to make a decision may not be available, may lack consistency and may not be expressed in a general way. How to retrieve and utilize data in design is an important problem. Two types of computer-based information systems that have been developed to manage product lifecycle and product-related data include product data management (PDM) and product lifecycle management (PLM). While promising, significant limitations still exist, where information required for decision making may not be available, may be lacking consistency, and may not be expressed in a general way for sharing between systems. The heterogeneity in the way of structuring and interpreting information causes conflicts, and makes it difficult to retrieve information from different sources. Wache [3] summarized those conflicts into structural and semantic conflicts. Structural conflicts arise because the same objects and facts in the domain can be described in different ways. Semantic conflicts occur when two systems do not use the same interpretation of information. Moreover, it is difficult for designers to consider multiple complex technical and economical criteria, relations, and objectives in product design.

In recent years, the concept of ontology has been used in the field of knowledge management and computer supported cooperative work (CSCW). Ontology is a formal specification of domain knowledge and has been used to define a set of data and their structure for experts to share information in a domain of interest. It is well suited for the representation and utilization of relations among data, and is efficient in knowledge reasoning. Ontology-based method is a new and promising approach to manage knowledge in engineering, integrate multiple data resources, and facilitate the consideration of the complex relations among concepts and slots in decision making. The purpose of this research is to explore the ontology-based method to solve the limitations in present computer-based information systems for product design.

There has been some research in ontology-based methods for information retrieval. Hwang, et al. [4] designed an architecture of search systems based on an ontology using web services. Since ontology can provide the inferred and associated information between data, in the system, ontologies are exploited by different semantic web applications to provide information to search clients. To search the increasing medical text documents, a medical information system, which has a semantic search function, was developed using a medical ontology that represents medical terminology semantic structure [5]. Vasilecas and Bugaite [6] put forward a method for ontology transformation into business rules, which are implemented by information-processing rules. In their research, the ontology axioms can be used to create a set of information-processing rules. They can be transformed into ECA rules and then to active DBMS triggers. Stojanovic N. and Stojanovic L. [7] presented a logic-based approach for query refinement in an ontology-based information portal. The query refinements are ranked according to their relevance to user's needs and have a self-improvement nature. In the ONTOWEB system [8], the ontology-based search engine is used to query the information that has been loaded into the database. This study showed that ontologies can be used not only to improve precision, but also to reduce the search time. While each of these works provides certain benefits, none focus on how to guide users to the proper information especially when designers are not very familiar with the process and when corresponding detailed data from different data resources belongs to multiple departments.

In this paper, a framework of ontology-based data integration and decision support for e-Design is built, using the hybrid approach in ontology-based integration. The framework can guide designers in the design process, with recommendations based on best historical data that meets previous inputs, and decision support based on knowledge and relations retained in ontologies. The sections that follow provide a detailed description of the approach, framework and associated ontology-based e-Design system for data integration and decision support.