📦 Logistics Performance Diagnostic: Central Asia Region

📋 Executive Summary

This project investigates a critical operational bottleneck: a 67% late delivery rate within the Central Asia region. By applying the Pareto Principle (80/20) and correlation analysis, I identified the "Vital Few" hotspots and provided strategic recommendations to optimize SLA compliance and restore brand reliability.

• Core Problem: High late delivery rates threatening customer trust and increasing operational overhead.

• Key Achievement: Isolated the top 20% of locations responsible for 80% of delays and debunked the "Volume Overload" myth through statistical verification.

🎓 Project Context: Self-Study & Research

This is a personal research project developed for self-learning purposes.

• Objective: To apply data analytics methodologies (Pareto, Correlation, EDA) to a real-world logistics dataset and practice end-to-end business problem-solving.

• Scope: The analysis is based on a publicly available dataset from Kaggle to simulate a business diagnostic process for educational and portfolio development.

• AI Collaboration: Developed with strategic assistance from AI tools for code optimization and documentation structure, ensuring rigorous statistical standards and professional reporting.

🛠️ Tech Stack & Methodology

• Language: Python (Pandas, Matplotlib, Seaborn).

• Analytical Techniques:

  • Pareto Principle (80/20 Rule): Used to eliminate statistical noise and focus on high-impact geographic hotspots.

  • Pearson Correlation: Evaluated the relationship between order volume and delivery performance to distinguish between capacity vs. process issues.

  • SLA Compliance Analysis: Benchmarked performance across different shipping tiers.

🔍 Key Insights

1. "First Class" Shipping Failure: Priority shipping modes exhibit delay rates comparable to standard modes, indicating a breakdown in the "Priority Lane" operational process.

2. Geographic Concentration: Identified specific states as hotspots where delays are localized, regardless of total order volume.

3. Seasonality Bottleneck: Delay rates spike significantly in Q4, revealing a lack of system elasticity during peak demand periods.

4. Operational vs. Capacity: Correlation analysis ($r \approx -0.15$) proved that delays are not driven by order volume, but rather by localized operational inefficiencies.

💡 Strategic Business Recommendations

1. Restructure "First Class" Workflow: Implement a dedicated priority sorting lane at regional hubs to ensure expedited orders meet their SLA commitments.

2. Resource Allocation for Hotspots: Redirect infrastructure investment or re-negotiate 3PL contracts specifically for the top 5 states identified in the Pareto analysis.

3. Peak Season Elasticity Plan: Scale temporary labor or expand transit warehouse capacity starting in early October to mitigate the Q4 seasonal surge.

4. Real-time Monitoring: Develop a triggered alert system for the Central Asia region when a state's late rate exceeds a 60% threshold.

🤝 Contributing & Feedback

I welcome any feedback, questions, or suggestions to improve this diagnostic model!

• Feedback: If you have insights on the methodology or business recommendations, please feel free to open an Issue or reach out via LinkedIn.

• Contribute: If you'd like to improve the code or visualizations, forks and Pull Requests are more than welcome.

📂 Repository Structure

• Analytics_Notebook.ipynb: Detailed Jupyter Notebook containing data cleaning, EDA, and statistical testing.

• data/: Directory containing the raw Kaggle dataset.

• images/: Exported visualizations (Pareto charts, Heatmaps, Trend lines).

👤 Contact Information

• Name: Phan Thi Thuy Anh

• LinkedIn: https://www.linkedin.com/in/blessed-thuy-anh/

• Email: thuyanhptta@gmail.com