Data Anonymisation in Industrial Big Data Systems

Digital industrial systems widely use big data and data-driven technologies to create digital twins, analyse, control, and apply artificial intelligence techniques in manufacturing. Data anonymisation is an emerging method for ensuring privacy in modern big data systems. The paper highlights the features of technological data that are important in terms of obfuscation (anonymisation) methods. The authors have analysed the methods of anonymisation problem formulation. The paper shows that the anonymisation of technological data can use their properties due to the state of the described data-driven environment, equipment and processes in indirect anonymisation. An example of using intermediate states based on Markov models in providing differential privacy of industrial data is given and its advantage is shown.