Thông tin chung
  English
  Đề tài NC khoa học
  Bài báo, báo cáo khoa học
  Hướng dẫn Sau đại học
  Sách và giáo trình
  Các học phần và môn giảng dạy
  Giải thưởng khoa học, Phát minh, sáng chế
  Khen thưởng
  Thông tin khác
  Tài liệu tham khảo
  Hiệu chỉnh
 
Số người truy cập: 107,382,978

 A deep learning approach for Control Chart Patterns (CCPs) prediction
Tác giả hoặc Nhóm tác giả: Tuan Le, Hai-Canh Vu, Nassim Boudaoud, Zohra Cherfi-Boulanger, Amelie Ponchet Durupt, Ho-Si-Hung Nguyen.
Nơi đăng: Research Publishing, Singapore; Số: 1;Từ->đến trang: 1267-1274;Năm: 2022
Lĩnh vực: Khoa học công nghệ; Loại: Bài báo khoa học; Thể loại: Quốc tế
TÓM TẮT
ABSTRACT
This paper presents a novel approach for predicting of control chart patterns (CCPs). The CCPs are used in Statistical Process Control to give information on the state of a process. There are six typical types of process states corresponding to six patterns: normal, cyclic, increasing trend, decreasing trend, upward shift, and downward shift. Except for the normal patterns, all the other patterns of control chart indicate that the process being monitored is not functioning correctly and requires adjustment. Knowing the system state or CCPs in advance and taking timely action helps reduce defective products and their incurred costs. In this study, the control chart time series are firstly created by joining the difference simulated CCPs with respect to the decreasing trend of their performance index (capability index). A deep learning (DL) model, convolutional neural network (CNN), is adapted for the CPPs prediction in the two following cases: (a) a time series contains only the normal patterns; (b) time series contains the six CCPs types. The model’s performance in both training and testing phases was assessed using mean absolute error (MAE). A parameter scanning of network parameters is conducted in order to gain information about the influence of the kernel size, number of filters, and dense size on CCPs prediction performance. The results show that the CNN model is able to not only capture the low and high levels of the CCPs quality variables, but also the mixed CCPs.
© Đại học Đà Nẵng
 
 
Địa chỉ: 41 Lê Duẩn Thành phố Đà Nẵng
Điện thoại: (84) 0236 3822 041 ; Email: dhdn@ac.udn.vn