PlumX Metrics
Embed PlumX Metrics

Neural Model of Manufacturing Process as a Way to Improve Predictability of Manufacturing

Lecture Notes in Mechanical Engineering, ISSN: 2195-4364, Page: 24-38
2022
  • 4
    Citations
  • 0
    Usage
  • 0
    Captures
  • 0
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Metrics Details

  • Citations
    4
    • Citation Indexes
      4

Conference Paper Description

The aim of this paper is to present the concept of a neural model of a manufacturing process. The goal of the model is to forecast the number of defective products based on the historical values of the manufacturing process parameters and the historical values of the number of defective products. The paper describes the creation of the model on the example of a manufacturing process in a glass factory. The use of the model allowed for a more accurate prediction of the number of defective products, which is treated as the improvement of the manufacturing predictability mentioned in the title of the paper. The model includes several NARX artificial neural networks. Each NARX network considers data from a different part of the production line. The forecast results on four test data sets are also presented. These results were compared with the classic approach, which uses a single neural network. The created model allowed for a significant reduction in the prediction error in four test data sets considered.

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know