Machine learning applications in production lines: A systematic literature review

Ziqiu Kang, Cagatay Catal*, Bedir Tekinerdogan

*Corresponding author for this work

Research output: Contribution to journalShort surveypeer-review

64 Citations (Scopus)


A production line is a set of sequential operations established in a factory where materials are put through a refining process to produce an end-product that is suitable for further usage. Monitoring production lines is essential to ensure that the targeted quality of the production process and the products are achieved. With the increased digitalization, lots of data can now be generated in the overall production line process. In parallel, the generated data sets are used by machine learning techniques for analytics of the production line to improve quality control, evaluate risks, and save cost. This paper aims to identify, assess, and synthesize the reported studies related to the application of machine learning in production lines, to provide a systematic overview of the current state-of-the-art and, as such, paving the way for further research. To this end, we have performed a Systematic Literature Review (SLR) in which we retrieved 271 papers, of which 39 primary studies were selected for a detailed analysis. This SLR presents and categorizes the production line problems addressed by machine learning, identifies the targeted industrial domains, discusses which machine learning algorithms have been used, and explains the adopted independent and dependent variables of the models. The study highlights the open problems that need to be solved and provides the identified research directions.

Original languageEnglish
Article number106773
JournalComputers and Industrial Engineering
Publication statusPublished - Nov 2020


  • Data analytics
  • Data mining
  • Machine learning
  • Production lines
  • Systematic literature review


Dive into the research topics of 'Machine learning applications in production lines: A systematic literature review'. Together they form a unique fingerprint.

Cite this