The rapidly changing environment of the shipbuilding industry has put Korea’s shipbuilding industry in a crisis. The purpose of this study was to develop a business model to maintain, maintain and operate Big Data-based MRO(Maintenance, Repair, and Operation) consumables, which is expected to be the new growth engine for the domestic shipbuilding industry. Although Korean shipbuilders have world-class technologies for ship dogma, the market for ship maintenance and repair is still in its infancy. For Korean shipbuilders, MRO business can be a growth engine that will provide food for the next 30 years, but to do so, we need to make sure that everything that happens in the entire process, from ship design to maintenance and maintenance. Therefore, by systematically establishing Big Data related to components and developing MRO business models based on data analysis capabilities using Artificial Intelligence system concept, we can develop new growth engines for related industries in Ship Industry.

The LOD data sets consist of RDF Triples based on the Ontology, a specification of existing facts, and by linking them to previously disclosed knowledge based on linked data principles. These structured LOD clouds form a large global data network, which provides a more accurate foundation for users to deliver the desired information. However, it is difficult to identify that, if the presence of the same object is identified differently across several LOD data sets, they are inherently identical. This is because objects with different URIs in the LOD datasets must be different and they must be closely examined for similarities in order to judge them as identical. The aim of this study is that the prosed model, RILE, evaluates similarity by comparing object values of existing specified predicates. After performing experiments with our model, we could check the improvement of the confidence level of the connection by extracting the link value.