Abstract

In the environment of industry 4.0, human beings are still an important influencing factor of efficiency and quality which are the core of product life cycle management. Hence, monitoring and analyzing humans' actions are essential. This paper proposes a vision sensor based method to evaluate the accuracy of operators' actions. Each action of operators is recognized in real time by a Convolutional Neural Network (CNN) based classification model in which hierarchical clustering is introduced to minimize the effects of action uncertainty. Warnings are triggered when incorrect actions occur in real time and applications of action analysis of workers on a reducer assembling line show the effectiveness of the proposed method. The research is expected to provide a guidance for operators to correct their actions to reduce the cost of quality defects and improve the efficiency of workforce.

Meeting Name

26th CIRP Conference on Life Cycle Engineering, LCE 2019 (2019: May 7-9, Lafayette, IN)

Department(s)

Engineering Management and Systems Engineering

Keywords and Phrases

Action recognition; Convolutional neural network; Hierarchical clustering; Real-time monitoring

International Standard Serial Number (ISSN)

2212-8271

Document Type

Article - Conference proceedings

Document Version

Final Version

File Type

text

Language(s)

English

Rights

© 2019 The Authors, All rights reserved.

Creative Commons Licensing

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

 
COinS