Leveraging synthetic data to tackle machine learning challenges in supply chains: challenges, methods, applications, and research opportunities

Yunbo Long, Sebastian Kroeger, Michael F. Zaeh, Alexandra Brintrup

Research output: Contribution to journalArticlepeer-review

Abstract

Machine learning (ML) has the potential to improve various supply chain management (SCM) tasks, namely demand forecasting, risk management, inventory management, production planning and control, supply network reconstruction, and distribution and logistics. However, the industrial application of ML in supply chains faces many challenges, particularly in data privacy and data scarcity. Synthetic data, which is artificially generated to mimic real-world data patterns, has shown promise in overcoming similar challenges in fields such as healthcare and finance. However, the application of synthetic data in the context of supply chains remains limited. This publication aims to analyze the challenges of machine learning operations (MLOps) for supply chain tasks and explain how synthetic data can address these challenges in a supply chain context. Moreover, the publication aims to identify suitable approaches to generate synthetic supply chain data. Based on the analysis, a research agenda is proposed as a guideline for future research activities to enable the use of synthetic data in the context of supply chains.

Original languageEnglish
JournalInternational Journal of Production Research
DOIs
StateAccepted/In press - 2025

Keywords

  • data generation
  • generative AI
  • Machine learning
  • MLOps
  • supply chain
  • synthetic data

Fingerprint

Dive into the research topics of 'Leveraging synthetic data to tackle machine learning challenges in supply chains: challenges, methods, applications, and research opportunities'. Together they form a unique fingerprint.

Cite this