Design of DNN-Based Injection Molded Product Defect Prediction System 


Vol. 46,  No. 10, pp. 1771-1777, Oct.  2021
10.7840/kics.2021.46.10.1771


PDF Full-Text
  Abstract

The government designated the injection molding industry as an additional field of the root industry by announcing the ‘Master Plan for Strengthening the Competitiveness of the Root Industry’ in July 2020. Injection molding is suitable for mass production systems for various types of products such as automobile parts and daily necessities, and is widely used to produce high-precision products. In this type of injection molding, the quality of the product is determined by various factors such as various working conditions of the injection machine, the temperature of the injection mold, and the dry state of the raw material, and the quality of this product is one of the very important factors in manufacturing management. Therefore, it is necessary to make efforts to minimize the occurrence of defective products in order to strengthen the manufacturing competitiveness of the injection industry. In this paper, data of various process variables of injection machine and mold are collected from IIOT, sensor, and PLC in real time, and types of good and defective products are determined through a DNN-based injection molding failure prediction model. We propose a system design that notifies the person in charge and the manager of the real-time analyzed defect information to take corrective action immediately.Through this study, it is expected that practical field improvement will be achieved by reducing the defective rate and non-operation rate, which are factors that decrease productivity at the manufacturing site.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Yeo and D. Park, "Design of DNN-Based Injection Molded Product Defect Prediction System," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 10, pp. 1771-1777, 2021. DOI: 10.7840/kics.2021.46.10.1771.

[ACM Style]

Seong-koo Yeo and Dea-woo Park. 2021. Design of DNN-Based Injection Molded Product Defect Prediction System. The Journal of Korean Institute of Communications and Information Sciences, 46, 10, (2021), 1771-1777. DOI: 10.7840/kics.2021.46.10.1771.

[KICS Style]

Seong-koo Yeo and Dea-woo Park, "Design of DNN-Based Injection Molded Product Defect Prediction System," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 10, pp. 1771-1777, 10. 2021. (https://doi.org/10.7840/kics.2021.46.10.1771)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880