A Methodology for Value Extraction in Industrial Alarms and Events
Alarms and events, Data Science, Industrial Big Data, Exploratory Data Analysis, Knowledge Discovery in Databases.
Alarm and event logs are an immense but latent source of information commonly undervalued in industry. Though, the current massive data-exchange, high efficiency and strong competitiveness landscape in manufacturing, boosted by Industry 4.0, does not accommodate such a data misuse and demands more incisive approaches when analyzing industrial data. Advances in Data Science and Big Data (or more precisely, Industrial Big Data) have been enabling novel approaches in data analysis which can be great allies in extracting hitherto hidden information from plant operation data. Coping with that, this doctoral qualifying work proposes the establishment of a data analysis pipeline especially designed to industrial alarm and event data, structured by a Knowledge Discovery in Databases (KDD) process and paved by current Data Science practices and methods. The design of an underlying Industrial Big Data infrastructure capable of meeting the large computational needs of the proposed approach is also in the scope of this work. Partial results on the proposal are shown in terms of an initial investigation based on Exploratory Data Analysis (EDA) of target data and the implementation of a prototype Industrial Big Data platform.