学术报告

Industrial Big Data: The Door to Prescriptive Analytics

讲座题目：Industrial Big Data: The Door to Prescriptive Analytics

主讲人：Prof. Diego Galar (Lulea University of Technology)

时 间：2016 年6月21日上午9:30

地 点：机械传动国家重点实验室 219 会议室

内容简介：

Assets are complex mixes of complex systems. Each system is built from components which, over time, may fail. When a component does fail, it is difficult to perform diagnosis or prognosis because the effects or problems that the failure has on the system are often neither obvious in terms of their source nor unique. Previous attempts to do diagnostics or prognostics of problems occurring in systems have been performed by experienced personnel with in-depth training and experience. However the big data era with computer-based systems collecting huge amount of data should overcome some of the disadvantages associated with relying on experienced personnel.

Moreover, the data collected are not only huge but often dispersed across independent systems that are difficult to access, fuse and mine due to disparate nature and granularity. That is why, rise of industrial data, its diversity and complexity over the past several years has been unprecedented. Data grows at an estimated exponential rate with no deceleration expected in the mid-term future. The current trend in data increase is already significantly affecting numerous areas of commercial, social and scientific domains but industries are special data repositories where data types, natures, and granularities dramatically increase. Expansion of industrial data notably outpaces our contemporary technological capacities to suitably process and manage it. Indeed, the rising discrepancy between the industrial data expansion and our technological means to cope with it highlights the pressing need for development of coordinated scientific approaches for predictive purposes but also for prescriptive ones.

In summary, there is a clear need to be able to quickly and efficiently determine the cause of failures and their propagation, propose optimum maintenance decisions, while minimizing the need for human intervention. For that purpose, there is a real need of massive data collection for such complex assets where much information needs to be captured and mined to assess the overall condition of the whole system and therefore the integration of asset information is required to get an accurate health assessment of the whole system, i.e. infrastructure, factories; facilities, vehicles etc.., and determine the probability of a shutdown or slowdown.

报告人简介：

Prof. Diego Galar holds a M.Sc. in Telecommunications and a PhD degree in Design and Manufacturing from the University of Saragossa. He has been Professor in several universities, including the University of Saragossa or the European University of Madrid, researcher in the Department of Design and Manufacturing Engineering in the University of Saragossa, researcher also in I3A, Institute for engineering research in Aragon, director of academic innovation and subsequently pro-vice-chancellor. He has authored more than three hundred journal and conference papers, books and technical reports in the field of maintenance, working also as member of editorial boards, scientific committees and chairing international journals and conferences. In industry, he has been technological director and CBM manager of international companies, and actively participated in national and international committees for standardization and R&D in the topics of reliability and maintenance. Currently, he is Professor of Condition Monitoring in the Division of Operation and Maintenance Engineering at LTU, Lulea University of Technology where he is coordinating several EU-FP7 projects related to different maintenance aspects, and was also involved in the SKF UTC centre located in Lulea focused in SMART bearings. He is also actively involved in national projects with the Swedish industry and also funded by Swedish national agencies like Vinnova. In the international arena, he has been visiting Professor in the Polytechnic of Braganza (Portugal), University of Valencia and NIU (USA), currently, University of Sunderland (UK) and University of Maryland (USA). He is also guest professor in the Pontificia Universidad Católica de Chile.