IP analytics for technology strategic decision making

Big data is increasingly available in all areas of manufacturing and operations. Increased data availability presents an opportunity for better decision making, to introduce the next generation of innovative and disruptive technologies. While intellectual property (IP) data is abundantly available, for many firms still remains a problem on how they can fully use this source of technical information. Firms struggle to decide how to better analyse IP data, to support and complement strategic decision making processes in the stages of technology and innovation development projects. In addition, while machine learning algorithms have widely been applied in other fields to analyse large amounts of data, they hardly been applied in the IP domain. We aim to complement technology strategic decision making with IP analytics, which in turn improves the human judgement at the technology development process. We follow a system design approach, where we design, develop and test a series of IP Decision Support Tools (DST), which make use of deep learning algorithms, to analyse patent data and classify a technology project, which is underpinned by a technology patent, as successful or not, to go through the innovation management funnel. From the literature, a successful patent can be defined as one with a large number of forward citations, one with consecutive renewal periods and one which has been litigated in court and won. The methodology is applied to a number of case studies to test its suitability within the technology development process. After refinement, we explore a number of alternatives to expand the model such as the addition of more data sources or its application at different stages of the innovation process. This methodology also improves the data quality, and the quality and validity of patents that are granted, as it benchmarks a potential application before the patent application stage. Project lead: Leonidas Aristodemou
Share This