Affiliations: [a] Department of Materials and Production, Aalborg University, Aalborg 9220, Denmark | [b] School of Informatics, Leicester University, Leicester LE1 7RH, UK
Correspondence:
[*]
Corresponding author: Chen Li, Department of Materials and Production, Aalborg University, Aalborg 9220, Denmark. Tel.: +45 52645576; E-mail: cl@mp.aau.dk.
Abstract: In light of recent trends toward introducing Artificial Intelligence (AI) to enhance the Human Machine Interface (HMI), companies need to identify the key issues of the communication between operator and production machines. Despite the fact that the industrial company starts to introduce chatbots to assist the communication between humans and machines, the virtual assistant (or digital assistant) by using human natural language is still widely required in the manufacturing domain. In this paper, we introduce an AI-based virtual assistant, Bot-X, for the manufacturing industry to handle a variety of complex services, e.g., order processing, production execution. This work expands the idea in three directions. Firstly, we introduce the design motivation of Bot-X, e.g., knowledge boundary in the manufacturing context. Secondly, the design principle of Bot-X is presented, including the framework, system architecture, model architecture, and the core algorithm. Then, three scenarios are presented to test the Bot-X usability and flexibility regarding the manufacturing environment.
Keywords: Intelligent manufacturing, Industry 4.0, virtual assistant, natural language processing
