A Review of Data-Centric Artificial Intelligence (DCAI) and its Impact on manufacturing Industry: Challenges, Limitations, and Future Directions

With the advancement of big data, the scope and potential of Artificial Intelligence (AI) have acquired major momentum. Data-Centric Artificial Intelligence (DCAI) is one of the most emergent fields of study in the current era of digitalization. Many examples have proven the effectiveness of Machine- and Deep Learning methods. In industrial production, however, limitations are still present that hinder application online and in a series that goes beyond isolated use cases. One crucial issue is data precondition, i.e., data quality, consistency, and labeling. As DCAI addresses these issues, developments in this field have caught the attention of various experts. In summary, DCAI continues to be an exciting and promising field of study that enhances AI applicability. Several research works have been conducted in DCAI, but unfortunately, no comprehensive reviews have been conducted to summarize and highlight the results. This gap in knowledge inspired our work, that aims to answer well-structured research questions. The focus of this paper is to clarify the terminology used in DCAI while also distinguishing it from other AI-related problems. This helps to analyze the current standards and problems associated with DCAI. This paper summarizes current use cases of DCAI and their impact on industries. Through this detailed description, readers can understand the potential and benefits of using DCAI in different business sectors. The analysis of the latest methods employed by DCAI to achieve enhanced AI performance and outcomes provides valuable insights for professionals and organizations that strive to incorporate AI into their business.