HarborFlow – Maritime GIS & AI Analysis

HarborFlow is a Maritime GIS & AI project focused on analyzing vessel traffic, port activity, weather conditions, and maintenance patterns.

It combines GIS mapping, Python analytics, SQL, and AI/ML to deliver operational insights and portfolio-ready outputs for the maritime sector.

---

🔍 What This Project Does

• 🚢 Analyze maritime traffic using AIS data
• 🌦️ Study weather impact on vessel behavior
• 🛠️ Simulate maintenance patterns for predictive analysis
• 🗺️ Visualize spatial data using GIS
• 🤖 Prepare data pipelines for AI & ML models
• 📘 Document a complete, professional workflow

---

📁 Repository at a Glance

HarborFlow_Maritime_GIS_AI_Analysis/ │ ├─ data/ → Raw, cleaned, and processed datasets ├─ docs/ → Technical logs & screenshots ├─ maps/ → GIS assets (lightweight only) ├─ notebooks/ → Jupyter analysis notebooks ├─ tableau/ → Dashboard outputs └─ README.md

Note: Large GIS project files (.aprx, .gdb) are intentionally excluded to keep the repository lightweight and portable.

---

📊 Project Status

Phase	Description	Status
Phase 0	GIS setup & base layers	✅ Completed
Phase 1	Data import & EDA	⚠️ In progress
Phase 2	Data cleaning & preprocessing	⏳ Planned
Phase 3	SQL analysis & KPIs	⏳ Planned
Phase 4	Visualizations & dashboards	⏳ Planned
Phase 5	Risk analysis & alerts	⏳ Planned
Phase 6	AI / ML modeling	⏳ Planned
Phase 7	Portfolio integration	⏳ Planned

---

🗺️ GIS Base Layers (Phase 0)

Loaded in ArcGIS Pro using Natural Earth (10m resolution):

• Ports: ne_10m_ports.shp
• Coastline: ne_10m_coastline.shp
• Land: ne_10m_land_scale_rank.shp

Note: The ArcGIS project file (.aprx) is not stored in the repo due to size and system dependency.
It can be shared via cloud storage on request.

---

🧪 Data Sources

Dataset	Description
AIS	Vessel traffic data (Port of Livorno)
Weather	Historical weather (Open-Meteo API)
Maintenance	Simulated proxy dataset

---

🛠️ Technologies Used

• GIS: ArcGIS Pro, Shapefiles
• Data: Python, Pandas, NumPy, SQL
• Visualization: GIS maps, Python plots, Tableau
• ML: Scikit-learn, Random Forest, XGBoost

---

📂 Detailed Documentation

• 🔹 Phase logs: docs/logs/
• 🔹 Screenshots: docs/screens/
• 🔹 Notebooks: notebooks/

Full technical details are documented separately to keep the repository clean and readable.

---

🚀 Next Steps

• Merge AIS, weather, and maintenance datasets
• Build SQL KPIs
• Develop predictive maintenance models
• Publish GIS & Tableau dashboards
• Integrate into professional portfolio