Hossain, Md Shahadat and Ali, Mohammad and Haider, Professor Parvin Sultana (2024) AI-Powered Predictive Analytics for Supply Chain Cyber Risk Management in Critical Infrastructure Industries. American Journal of Technology Advancement, 1 (1). ISSN 2997-9382

Text AJTA_AI-Powered+Predictive+Analytics.pdf Download (551kB)

Abstract

Today's software development relies on increasingly interconnected applications, third party components and open-source dependencies, creating software supply chain attacks as a major threat to cyber security. The present study explores the use of AI powered predictive analytics in Software Supply Chain Cyber Risk Management in the context of the Software Supply Chain Security Dataset. The goal of the research is to investigate the cyber incidents in software supply chain, find out the important cyber risk factors, build predictive models on Decision Tree, Random Forest and Naive Bayes algorithms and assess the effectiveness of using artificial intelligence in predicting cyber risks. A quantitative research design was used, and the data was preprocessed, followed by exploratory data analysis, classification using machine learning techniques, feature importance analysis, and predictive modelling with R Studio. The results show that the Distribution Vector, Attack Vector, Codebase and Attacker Type are the most significant factors that influence software supply chain cyber risks. The model Decision Tree was able to classify cyber incidents effectively, the model Random Forest was able to identify the most important predictive variables and the model Naive Bayes was able to be used to successfully predict the cyber risk level for several category types of incidents. The study illustrates how AI-driven predictive analytics can help make more informed decisions in proactive cybersecurity management, enhance cyber risk identification, prediction, and strategic planning, and foster more resilient and secure software supply chain ecosystems.

Item Type:	Article
Uncontrolled Keywords:	Software, AI-Powered, Analytics, Supply Chain Security, Dataset, Decision Tree, Random Forest, Naive Bayes, Algorithm, Cyber Risk
Subjects:	H Social Sciences
Depositing User:	admin eprints
Date Deposited:	30 Jul 2026 15:46
Last Modified:	30 Jul 2026 15:46
URI:	http://eprints.umsida.ac.id/id/eprint/16856

Actions (login required)

View Item