Supply Chain Delay Prediction System

Overview

This project predicts shipment delay risk and provides actionable recommendations to improve delivery performance.

Problem

Delays in supply chain operations affect efficiency and customer satisfaction. The goal is to identify high-risk shipments early and support better decisions.

Solution

• Built a machine learning model to predict delay risk
• Engineered features such as delivery duration and scheduling gaps
• Deployed the model as a Streamlit application
• Added automated recommendations based on risk level
• Implemented a what-if simulator for scenario testing

Key Features

• Delay prediction (Low, Medium, High risk)
• Automated decision support
• Scenario simulation
• Downloadable report

Screenshot

Live App

https://supply-chain-delay-app-nzckm7ysyhpybwd7uqwojj.streamlit.app/

Tech Stack

• Python
• Pandas, Scikit-learn
• Streamlit

Project Structure

• app.py → Streamlit application
• notebook/ → data analysis and modeling
• requirements.txt → dependencies

Key Insight

Delivery delays are driven more by operational factors such as delivery duration and scheduling gaps than external conditions in this dataset.

Future Improvements

• Separate training and inference for production deployment
• Integrate real-time data
• Deploy as an API