Supervisor

Dr. Taufique Ahmed

Programme

MSc in Data Analytics

Subject

Computer Science

Abstract

Supply chain management has always been a challenging area where delivery delays stand as one of its main ongoing issues. Many companies continue to use outdated data collection systems despite the advancements in data collection tools. This study investigates how predictive analytics can help companies to prevent their operational problems from reaching their customers.

In this study, two datasets were used. The first is the Data Co Smart Supply Chain dataset which covers order level operations from 2015 to 2018, and the second is the World Bank Logistics Performance Index, which covers logistics quality across more than 160 countries. After the combination and cleaned of both datasets, seven machine learning models were evaluated and tested to predict delivery delays. Then an LSTM deep learning model was developed to predict average daily delivery delays for the upcoming 15 days.

The results showed that the type of shipping and the scheduled delivery time were the two important factors in deciding whether an order arrives late or not. The Stacking Classifier model tested in this research achieved the highest performance result with a ROC-AUC score of 0.733, but the simpler models showed performance results that were very close to that level of performance. The LSTM model operated at its highest performance level when the Ada max optimiser was used for training, which enabled it to produce accurate short-term delay predictions that could assist planning activities. This research demonstrates that predictive analytics provides significant benefits to logistics operations. This system provides managers with advance notice of upcoming events which allows them to take appropriate actions at the correct time.

Date of Award

2026

Full Publication Date

2026

Access Rights

open access

Document Type

Capstone Project

Resource Type

thesis

Download

Included in

Data Science Commons

Share

COinS