Industry 4.0 Vision for the Supply of Energy and Materials

5 Applications of Artificial Intelligence and Big Data in Industry 4.0 Technologies

Reza Rezazadegan and Mahdi Sharifzadeh^*

Sharif Energy, Water and Environment Institute (SEWEI), Sharif University of Technology, Tehran, Iran* Corresponding author. Sharif Energy, Water and Environment Institute (SEWEI), Sharif University of Technology, Tehran, Iran

5.1 Introducing Artificial Intelligence (AI)

AI can be described as enabling machines to do tasks that humans can perform [12]. Unlike natural sciences or mathematics in which problems are solved based on fundamental laws or axioms, in machine learning we attempt to solve problems using heuristic and probabilistic algorithms. There are different ways to classify AI methods. On a first note, AI can be divided into symbolic AI and machine learning (ML). Symbolic AI, which belongs to the first wave of artificial intelligence, tries to encode human knowledge into machines using symbolic logic. One tries to emulate intelligence by mechanistic manipulation of the symbols. This way, AI is regarded as modeling the relationship between symbolic variables and structures [13]. Expert systems, which mimic the thinking of a human expert such as a doctor, are an example of symbolic AI. On the other hand, ML, which is responsible for the current AI and data science boom, tries to learn numerically from a pool of labeled or unlabeled data.

ML is already widely used in industry [7]. Its applications include fault detection, maintenance, decision ...