Challenges and Frontiers in Implementing Artificial Intelligence in Process Industry

The implementation of artificial intelligence faces different challenges of infrastructural, data related, security related and social scope. These aspects are discussed, reflecting on the requirements of introducing such a technology in a broader way. Machine learning, as subfield of artificial intelligence, benefits from advances in Big Data science. One example is the $$\lambda $$ -architecture which can be used for the treatment of streamed process data in industrial applications. Digital twins are shown as further tool providing object-oriented data for machine learning applications. Yet, the increasing freedom of data transfer within a plant, as propagated by Industry 4.0, poses new risks for information technology and automation systems: security of those components is one of the big challenges. Here, artificial intelligence can seen as both risk and solution. A last relevant challenge is acceptance among the staff, as artificial intelligence is associated with fears. Counterstrategies for those fears are presented as a proposed guideline for real applications. Finally, current frontiers at process industry are considered and discussed. These include the need for strengthening the use of high-dimensional data availability, increased roll-out of optimisation concepts and rigorous progresses in semantic modelling of processes and process chains, in order to fully exploit the beneficial scope of artificial intelligence in industry.