Optimising Supply Chain Logistics with Predictive Analytics

Bhambid, Sidhesh Sakharam (2025) Optimising Supply Chain Logistics with Predictive Analytics. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (749kB) | Preview

Abstract

Modern supply chains operate within a highly volatile and complex global environment, where disruptions can lead to significant financial losses and damage to brand reputation. The ability to proactively identify and mitigate risks is a critical competitive advantage. This research project addresses the challenge of optimising supply chain logistics through the application of predictive analytics. The primary objective is to develop and evaluate a machine learning model capable of accurately classifying the risk level of logistics operations. The study utilises a comprehensive dataset encompassing over 32,000 logistical events, detailed with 26 features including real time vehicle tracking, fuel consumption, inventory levels, and environmental factors. Following a rigorous methodology of data preprocessing, feature engineering, and dimensionality reduction using Principal Component Analysis (PCA), a comparative analysis of five supervised learning algorithms was conducted: Decision Tree, AdaBoost, K Nearest Neighbors, Gaussian Naive Bayes, and Support Vector Machine (SVM). After careful hyperparameter tuning and using evaluation metrics including accuracy, precision, recall, F1-score, and ROC-AUC, the SVM model was the best predictor of the classes. The final retrained SVM model, after balancing classes and engineered features, had amazing accuracy of 97.83%, with an F1 score of 97.87% on test data. The results of this study highlight how effective predictive analytics can evolve logistics management from a reactive process to one that is proactive, providing a strong instrument to help inform decisions, reduce costs, and build the resilience of supply chains.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Milosavljevic, Vladimir UNSPECIFIED
Subjects:	H Social Sciences > HF Commerce H Social Sciences > HC Economic History and Conditions > Globalisation H Social Sciences > HD Industries. Land use. Labor > Business Logistics > Supply Chain Management
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2026 11:22
Last Modified:	25 Aug 2026 11:22
URI:	https://norma.ncirl.ie/id/eprint/9621

Actions (login required)

View Item