A dynamic information schema for supporting product lifecycle management C.K.M. Lee * , G.T.S. Ho, H.C.W. Lau, K.M. Yu Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hunghom, Hong Kong Abstract The globalization of competition and changing customer requirements are the emerging phenomenon of today’s manufacturing environment. Nowadays, manufacturers need useful information prior to making important decisions based on their experience and knowledge. However, the lack of seamless data transformation between database and knowledge based system hinders design practitioners from responding rapidly to the changes in the complex and dynamic product development environment. This paper proposes a dynamic product information schema which is characterized by its ability to provide designers with a product data exchange standard, thus, enhancing the responsiveness of product development. The approach is based on the modification of the XML standard which is assimilated with object technology to form a cross- platform protocol that enables enterprises to update their data constantly. In particular, data from the traditional relational database model can be transformed to a format which is acceptable in a knowledge base system. To validate the feasibility of the proposed schema, a trial run has been conducted in a local company based on the suggested approach. q 2005 Elsevier Ltd. All rights reserved. Keywords: Object technology; Product development; Product information markup language (PIML); Extensible markup language (XML); Extensible stylesheet language transformations (XSLT) 1. Introduction Due to the keen competition and short time-to-market, manufacturers need to leverage their knowledge and keep their value-added processes to strengthen their core business. Product design, which is recognized as a critical process, determines 70% of the product cost. In general, product design is an evolving process of problem solving that consists of acquiring requirement, creating product concept, studying feasibility to be manufactured with existing technology and finally providing the overall artifact solution. Manufacturers have to face the problem of unclear customer requirements, short delivery time and technical constraints. Quality, cost and time, which are the three main dimensions for measuring the product development process, are often a matter for discussion (Bullinger, Warschat, & Fisher, 2000; Kengpol & O’Brien, 2001). To achieve the above mission, the principle of Product Lifecycle Manage- ment (PLM) emerges as a concept within operation management which is concerned with the process of designing product and production system as well as the realization of production, distribution and service. Data is achieved, accessed, manipulated within the product develop- ment lifecycle such that Product Data Management (PDM) enables the integration of information generated by CAD/ CAM and CAPP. As a result, the benefits of PDM, which are information sharing, workflow optimization and product structure management, helps pave the way of concurrent engineering. Intelligent data analysis with hypothesis enables knowledge discovery and knowledge acquisition, both of which can be achieved by systematic formulation of the meaningful information through reasoning. To enable the proper use of captured knowledge, it is important that available data can be organized to information which is then systemized to form knowledge, thus, allowing the reuse of expertise advice in an efficient way. In particular, this paper explores the potential of deploying XML as universal standard for product data interchange between case base and database. It proposes a data exchange method which consists of a set of XML capitalized with the feature of object technology (such as polymorphism and inheritance). Product Information Markup Language (PIML) is developed in OBKIS to access and manipulate product data in geographical dispersed environment. Within the proposed system, CBR provides the mechanism that retrieves the past similar cases and reuses them in the new problem situation, thereby achieving a certain level of incremental and sustained Expert Systems with Applications 31 (2006) 30–40 www.elsevier.com/locate/eswa 0957-4174/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2005.09.009 * Corresponding author. Tel.: C852 2766 6628; fax: C852 2362 5267. E-mail addresses: mfrsch24@inet.polyu.edu.hk (C.K.M. Lee), mfhenry@ inet.polyu.edu.hk (G.T.S. Ho), mfhenry@inet.polyu.edu.hk (H.C.W. Lau), mfkmyu@inet.polyu.edu.hk (K.M. Yu).