Privacy-Preserving Techniques for Machine Learning Applications in Supply Chains

Privacy-Preserving Techniques for Machine Learning Applications in Supply Chains

Joshi, Ayush (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Intelligent Systems; TU Delft Cyber Security)

Li, T. (mentor)
Erkin, Z. (mentor)
Hildebrandt, K.A. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2021-07-01

Supply chains are vital to the global economy, and so, increasing efficiency in supply chain management is of utmost importance. Modernizing technology has allowed for various uses of machine learning to be possible in several aspects of supply chains, specifically in demand forecasting with prediction models, and customer relations with chat-bots. While this may be the case, many organizations are reluctant to implement such solutions due to potential threats to their privacy. In addition to this, some currently existing solutions do not take special care for privacy preservation. This brings the question of, "How can privacy be preserved in machine learning based applications in supply chains?" The results of this survey show that several approaches for privacy-preservation of machine learning applications exist, and can be applied to supply chains while maintaining increased efficiency in supply chain management.

machine learning
privacy
supply chains
demand forecasting
Chatbot
Privacy Preserving Machine Learning
privacy preservation

http://resolver.tudelft.nl/uuid:9c443417-09aa-4f96-8a6f-f6d8f686e1f1

Student theses

bachelor thesis

© 2021 Ayush Joshi