Ontology-Enhanced Life Cycle Assessment: A Case Study of Application in Oil Refinery Akkharawoot Takhom † Boontawee Suntisrivaraporn † Thepchai Supnithi ‡ † School of Information, Computer and Communication Technology, Sirindhorn International Institute of Technology, Thammasat University, Thailand ‡ Language and Semantic Technology Laboratory, Intelligent Informatics Research Unit, Intelligent Informatics Research Unit,National Electronics and Computer Technology Center, Thailand akkharawoot.t@gmail.com Abstract This paper presents an improvement of a pre- vious work on an ontology for life cycle assess- ment or O-LCA by taking into consideration the life cycle inventory (LCI) and life cycle impact assessment (LCIA). The knowledge about en- vironment, manufacturing processes, participat- ing substances, as well as that about assessment methods, are represented in the web ontology language. To demonstrate the benefits of em- ploying a logical formalism and inference in life cycle assessment, we implement a resource rec- ommender system built on top of our developed ontology, and carry out a usability study of the system. With respect to a case study of oil re- finery, the evaluation suggests that though there are still a number of requirements from the field users, the system was proven useful in term of user satisfaction. 1 Introduction To measure an environmental impact, a kind of environmental impact assessment (EIA) tools called Life Cycle Assessment (LCA) is used to identify, quantify energy and materials which used and released to the environment, and eval- uate and implement opportunities to influence environmental improvements. To utilize the knowledge, the Web Ontology Language (OWL) [1], endorsed by the World Wide Web Consor- tium (W3C), is applied to improve the knowl- edge domain, called O-LCA ontology [2]. The Description Logic (DL) [3] manifests its capabil- ity to formalize the LCA ontology, allowing us to represent and reason with conceptual knowledge [4]. Automatic reasoning of DL is tractable as it is a tractable subset of first-order predicate logic, resulting in expressivity and robust scalability. In previous work of LCA knowledge and se- mantic web technology, the standard guideline has been used to interpret and adapt with the specific objectives. Some of related work fo- cuses on utilization of different standard guide- lines. For example, CASCADE [5] project aims to adapt the data document format with other standard guidelines. LCAO [6] is another project that interprets the LCA framework, and aims at the organization and retrieval of information, and at the contribution for a consensual vision. In the case study of U.S. energy impact data man- agement [7]. LCI can also be semantically rep- resented as manipulatable databases using rela- tional algebra. In this approach an ontology is used to model associate elementary processes in LCI according to their interdependency rela- tions. For enhancing the formalized LCA knowl- edge, this work carries on improving the O LCA [2] by expanding LCA phase to Impact Assess- ment (LCIA). The renovated knowledge is im- plemented as a recommender system by O LCA Framework and applied to a case study of oil re- finery. By working proof, an inferential ability of ontology based on DLs system is utilized in the recommender system that follows the approach of Cleaner Technology [8] for environmental im- pact reductions. The remainder of this paper is as follow. We will start with background knowledge in both of knowledge representation and Life Cycle As- sessment as the domain knowledge. Next sec- tion, an implementation with O LCA Framework will be described: redesigning enhanced the on- tology to support the framework, and remodel- ing ontology from exists to a new O LCA . A case study of oil refinery is chosen and adjusted to be suitable usage in the fourth section. With a pur- pose of the practical usage, important issues will be discussed in the fifth section. Finally, we will