Development of an Ensemble Modeling Framework for Data Analytics in Supply Chain Management

Chibuzor Joseph Udokwu, Robert Zimmermann, Patrick Brandtner, Tobechi Michael Obinwanne

Research output: Contribution to journalArticlepeer-review

Abstract

The application of Data Analytics (DA) in Supply Chain Management (SCM) provides a plethora of opportunities for improving the performance, efficiency, and resilience of organizations by predicting the behaviour of the supply chain network. In this regard, the use of complex DA approaches that involve the combination of multiple Machine Learning (ML) models such as the ensemble approaches have shown to be highly effective in improving prediction accuracy and model performance. Still, the application of these ensemble approaches has not yet been properly explored in SCM analytics. Thus, this paper presents an ensemble model framework specifically designed for SCM issues by exploring problem types and use cases where ensemble approaches can be applied in SCM. The developed framework enables the selection of the right ensemble models for specific SCM DA tasks. Thus, this paper contributes to the scientific body of knowledge by presenting a systematic approach to ensemble modelling in SCM analytics by outlining suitable ensemble model compositions for different SCM use cases and data analytics problems. Hence the results of this paper provide a potential tool for DA practitioners in SCM to further improve the performance of their Supply Chain (SC) networks by selecting appropriate ensemble elements for different SCM analytics problems.

Original languageEnglish (American)
Pages (from-to)1289-1300
Number of pages12
JournalJournal of Advances in Information Technology
Volume14
Issue number6
DOIs
Publication statusPublished - 27 Nov 2023

Keywords

  • consensus algorithm
  • data analytics
  • ensemble model
  • machine learning
  • predictive analytics
  • supply chain analytics

Fingerprint

Dive into the research topics of 'Development of an Ensemble Modeling Framework for Data Analytics in Supply Chain Management'. Together they form a unique fingerprint.

Cite this