User:Noor Nadhiya Mohammad

Abstract
Clearly defined product structure is a key foundation for product manufacturing requirements. Automated information extraction of the 2D CAD engineering drawing ensures a more precise and accurate extracted information, making it the best way to provide product manufacturing requirements compared to reading and interpreting the drawing manually. However, the occurring problem from the automation process is the existing of anonymous terms due to heterogeneity in engineering drawing. Besides that, heterogeneous of product in engineering manufacturing might cause product structure data to be represented in different ways, adding to the complexity of product configuration. Therefore, to solve the problem of heterogeneity of terms and products, we proposed a dynamic ontology called Product Structure Ontology (PSO) in which the vocabulary can be controlled and standardized, making it suitable to be applied to all automotive spring products. Here, we introduced the methodology used to develop PSO comprehensively by providing artifacts such as category, anatomy and scheme. With these artifacts, the PSO can be extended, reused and even duplicated into other products. The overview of the PSO annotation and its evidence are also presented. We also provide PSO resources including PSO website, browser, database availability and documentation.

Introduction
There has been much discussion in recent years about formalizing knowledge using ontology whether for general usage or specific domains. Noy and McGuiness (2001) definition affirm that ontology defines a common vocabulary for researcher, who needs to share information in a domain of use. For more appropriate meaning, ontology is a category of things that exist or may exist in some domain. It is an explicit formal specification of the objects, concepts and other entities that are assumed to exist in some area of interest and the relationships that hold among them. In the context of Artificial Intelligence, we can describe ontology by defining a set of representational terms. As there are many concepts in product manufacturing requirements and it remains divergent among company, ontology becomes important to support the sharing and reuse of formally represented knowledge by explicitly stating concept, relations and axioms in this domain. The best way to provide product manufacturing requirements is by information extraction of the 2D CAD engineering drawing compared to reading and interpreting the drawing manually. The information extraction for example involves table extraction and process identification to capture elements contained within 2D CAD engineering drawing table, geometric objects and its dimension respectively. By doing this, tendency to miss such indications can be avoided, thus information extracted is more precise and accurate. Also, inconsistencies in product manufacturing requirements caused by varying engineer’s experience and knowledge can be avoided. However, the occurring problem from the automation process is the existing of anonymous terms due to heterogeneity in engineering drawing. In this research, we are more focused on engineering terms used in 2D CAD engineering drawing and leaf spring product design in automotive manufacturing industry as our case study. Product Ontology (PRONTO) by Marcela et al. (2005) defines concept, relation among them and axioms to be applied in the complex product modeling domain. It is primarily related with complex product structure which involves an association of hybrid structures (combining composition and decomposition types of operation) to end product like in some food (milk and meet) and petrochemical industries. Another work which engages ontology in order to represent product was discussed by Lee et al. (2006). The study reports the effort to build an operational product ontology system for government procurement services which is designed to serve as a product ontology knowledge base. Thevenot et al. (2006) presents their approach to retrieve and reuse relevant information when redesigning product. They used heuristics and shared ontological component information and proposed a framework to capture, store, retrieve, reuse and represent information for product family redesign using ontology, graph query, formal concept analysis, commonality assessment and genetic algorithm-based optimizer. As a reference to utilize ontology in manufacturing field, Manufacturing’s Semantic Ontology (MASON) proposed by Lemaignan et al. (2006) aimed to draft a common semantic net in this domain. Their paper discussed a usefulness of ontology for data formalization and sharing, and shows the sufficiency of Web Ontology Language (OWL) for ontologies especially in manufacturing environment. Meanwhile, Lepratti (2006) proposed an innovative ontology-based approach called Ontological Filtering System (OFS) to formalize natural language contents in a systematic way. They present a concept for improving the human-machine interaction by means of an innovative procedure for the processing of natural language instruction. Although ontology are now in widespread use, but the nature and the use of ontologies are unfamiliar. Therefore, Darlington and Culley (2008) takes a practical approach through the use of example to clarify what ontology is and how it might be useful in an important and representative phase of the engineering design process. They discussed the use of ontology and explored a methodology for developing ontology. They also described the application of these ontologies for supporting the capture of the engineering design requirements. From the literature review we gained insight of ontology concept and its use. These authors however, did not release their ontology design and consequently the developed ontology cannot be reused or duplicated to other products. To solve the problem of heterogeneity of terms and products, we come out with a dynamic ontology called Product Structure Ontology (PSO) in which the vocabulary can be controlled and standardized, making it suitable to be applied to all automotive spring products. Our methodology for developing the PSO is based on Noy and McGuinness methodology which has been modified based on its suitability. We also come out with two PSO artifacts called PSO Category and PSO Anatomy. Whereby, with these artifacts, the PSO can be extended and reused, duplicated into other product as well as its annotation being made in comprehensive and complete manners.

Education

 * 2010, M.Sc Computer Science
 * 2008, BSc Computer Science (Management Information System), Universiti Teknologi Malaysia.
 * 2005, Dip Computer Science (Information Technology), Universiti Teknologi Malaysia.

Research Interests

 * 1) Product Manufacturing Requirements.
 * 2) 2D CAD Engineering Drawing and Product Structure.
 * 3) Knowledge Formalization.
 * 4) Ontology.
 * 5) 2D CAD Engineering Drawing Extraction Algorithm.

Group Members

 * 1) Ismail M. Amin, Supervisor.
 * 2) Razib M. Othman, Researcher.
 * 3) Hishammuddin Asmuni, Researcher.
 * 4) Shafry M. Rahim, Researcher.
 * 5) Faizal M. Jabal, MSc Student (2D CAD Engineering Drawing Extraction).
 * 6) Khairuddin M. Ahamad, MSc Student (Multi-Agent System for 2D Engineering Drawing Extraction).
 * 7) Azri A. Majid, Programmer.
 * 8) Asarudin M. Tap, Programmer.
 * 9) Noraishah S. A. Rahman, Programmer.
 * 10) Syazrah N. Z. Othman, Proof-reader.
 * 11) Rohayanti Hassan, Technical Writer.
 * 12) Harun Yaacob, Web Designer.
 * 13) Husni Rulai, Web Designer.

Contact Info
Nadhiya N. Mohammad Laboratory of Computational Intelligence and Biology (LCIB) No. 204, Level 2 Industry Centre, Technovation Park Universiti Teknologi Malaysia (UTM) Jalan Pontian Lama 81300 Skudai, Johor, MALAYSIA Mobile: +6014-802-8526 Tel/Fax: +607-559-9230 E-mail: [mailto:nadhiyamohammad@gmail.com nadhiyamohammad@gmail.com]