A Prediction Method of Order Completion Time Based on BP-SVR

In this paper, an order completion time prediction method based on back propagation neural network (BP) - support vector regression (BP-SVR) is proposed for multi product and multi machine shop manufacturing environment. After identifying the factors affecting the completion time, firstly, BP is app...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu, Jiahui, Li, Tieke, Wang, Bailin, Zhang, Wenxin, Zhang, Zhuolun
Format: Conference Proceeding
Language:English
Subjects:
BP neural network
Computational modeling
Data models
Feature extraction
Manufacturing processes
Neural networks
order completion time
Predictive models
simulation model
Support vector machines
support vector regression
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, an order completion time prediction method based on back propagation neural network (BP) - support vector regression (BP-SVR) is proposed for multi product and multi machine shop manufacturing environment. After identifying the factors affecting the completion time, firstly, BP is applied as a feature extractor to the simulation data set containing the target features. Secondly, the extraction results are input into SVR for prediction, and finally the evaluation indicators are determined and experimentally verified. The results show that the BP-SVR model performs better in terms of goodness of fit and prediction error, and can obtain ideal prediction results.
ISSN:2473-3547
DOI:10.1109/ISCID56505.2022.00017