---

Delivery Intelligence System

Team Members

• Faustine Emmanuel
• Cleven Godson

Problem

Delivery operations can often be unpredictable, leading to delays, inefficiencies, or even fraud. Understanding the factors affecting delivery times and identifying anomalies can significantly improve decision-making, reduce costs, and enhance customer satisfaction. However, tracking and analyzing such data manually can be time-consuming and prone to errors.

Solution Overview

The Delivery Intelligence System is a smart solution designed to streamline delivery operations using AI and data analytics. Built as a Flask web application, it helps delivery companies predict delivery times and detect anomalies in real-time. This system processes CSV datasets containing delivery information, providing actionable insights through a user-friendly interactive dashboard.

The key features include:

• Prediction of delivery times using machine learning.
• Anomaly detection to identify unusual delivery patterns that could indicate fraud or operational issues.
• Interactive visualizations to help users easily understand the data and make informed decisions.

Instructions to Run the Project

To run the Delivery Intelligence System locally, follow these simple steps:

1. Clone the Repository

First, clone the repository to your local machine:

git clone https://github.com/cleven12/delivery_intel_system.git
cd delivery_intel_system

2. Install Dependencies

Make sure you have Python 3.x installed. Then, install the required libraries:

pip install -r requirements.txt

3. Run the Flask Application

Start the Flask application by running:

cd src
python app.py

This will start a local web server. You can access the application by visiting http://127.0.0.1:5000/ in your browser.

4. Use the Application

Once the app is running, you can:

• Upload CSV files: Click the "Upload File" button on the homepage to load your delivery dataset.
• View the dashboard: The dashboard will display predictions, anomalies, and analytics for your data.

Special Requirements

• Python Version: Python 3.6 or higher.

• Libraries:

  • Flask
  • Pandas
  • NumPy
  • Scikit-learn
  • Chart.js (for visualizations)
  • Bootstrap (for the UI)

You can install all the dependencies by running:

pip install -r requirements.txt

• File Handling: The system handles CSV file uploads up to 200MB. Ensure your dataset is in CSV format and has proper columns (e.g., delivery times, city names, timestamps).

Deployment

You can access the live version of the system at the following URL: Kilitech Duo - Delivery Intelligence System

The code for this project is available on GitHub: GitHub Repository

System Architecture

Frontend Architecture

• Responsive UI: Built using Jinja2 templates and Bootstrap, ensuring a mobile-friendly, sleek design with a dark theme.
• Interactive Charts: Data visualizations powered by Chart.js, including real-time analytics for delivery predictions and anomaly detection.
• Single-Page Application Flow: Upload → Processing → Dashboard, with persistent session data.

Backend Architecture

• Flask Web Framework: Simple and lightweight Python server that handles all web requests and serves the dashboard.

• Modular Design:

  • app.py: Manages routing and requests.
  • data_processor.py: Handles data cleaning and preprocessing.
  • ml_models.py: Contains the machine learning models for predictions and anomaly detection.

• Session Data: Uses in-memory session storage for temporary data during the demo.

Machine Learning Architecture

• Prediction Model: A custom linear regression model to predict delivery times using NumPy.
• Anomaly Detection: Uses statistical methods (isolation techniques) to detect unusual delivery patterns.
• Training Pipeline: Data cleaning → Feature extraction → Model training → Evaluation.

Data Processing Pipeline

• CSV Ingestion: Uses Pandas to load and process delivery datasets.
• Data Cleaning: Standardizes timestamps, handles missing data, and detects outliers.
• Feature Engineering: Extracts features like delivery durations and normalizes numerical columns.
• Validation: Ensures data quality with error handling during the process.

External Dependencies

Core Web Framework

• Flask: The backbone of the web application.
• Werkzeug: Provides utilities like secure filename handling and middleware support.

Data Science Stack

• Pandas: Used for CSV processing and data manipulation.
• NumPy: For numerical operations and custom machine learning models.
• Scikit-learn: Provides metrics for model evaluation.

Frontend Libraries

• Bootstrap: A front-end framework for responsive design and mobile-friendly UI.
• Chart.js: For creating interactive visualizations of data like predicted vs actual delivery times.
• Font Awesome: Enhances the UI with icons.

Development Tools

• Python Logging: Built-in logging for tracking errors and debugging.
• Werkzeug ProxyFix: Middleware for header handling during deployment.

File System Dependencies

• OS Module: For environment variable management and directory handling.
• Uploads Directory: Used for storing CSV files temporarily during processing.

---