A neural network approach to predict deliverability in manufacturing

Zhen Hao, Ahmed Abukar*, Mozafar Saadat, Salman Saeidlou

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

At present, estimating the deliverables of products in the manufacturing industry mainly depends on the knowledge of experts, but the knowledge of experts is sometimes difficult to obtain and often limited. The main purpose of this paper is to provide a method that can predict the deliverability of a product based on machine learning and data science technologies. The data used is real industry data and the machine learning algorithm used does fit the data set. The results of the model illustrate that using machine learning algorithms to predict deliverability is feasible. The machine learning algorithm achieves precision and recall percentages respectively. However, the findings also address some limitations. The machine learning algorithm has requirements on the form of the data. If the new data and the historical data have different forms, the machine learning algorithm cannot generalize well on new data.

Original languageEnglish
Title of host publicationAdvances in Manufacturing Technology XXXII - Proceedings of the 16th International Conference on Manufacturing Research, ICMR 2018, incorporating the 33rd National Conference on Manufacturing Research
EditorsKeith Case, Peter Thorvald
PublisherIOS Press BV
Pages361-366
Number of pages6
ISBN (Electronic)9781614994398
DOIs
Publication statusPublished - 2018
Event16th International Conference on Manufacturing Research, ICMR 2018 - Skovde, Sweden
Duration: 11 Sept 201813 Sept 2018

Publication series

NameAdvances in Transdisciplinary Engineering
Volume8

Conference

Conference16th International Conference on Manufacturing Research, ICMR 2018
Country/TerritorySweden
CitySkovde
Period11/09/1813/09/18

Keywords

  • Machine Learning
  • Neutral Network
  • Predictive Scheduling

ASJC Scopus subject areas

  • Computer Science Applications
  • Industrial and Manufacturing Engineering
  • Software
  • Algebra and Number Theory
  • Strategy and Management

Fingerprint

Dive into the research topics of 'A neural network approach to predict deliverability in manufacturing'. Together they form a unique fingerprint.

Cite this