🚢 Pakistan Maritime Trade & Shipping Dataset (2020–2026)
📌 About the Dataset
This synthetic dataset simulates maritime trade and shipping activity in Pakistan from January 2020 to April 2026. It is inspired by the operations of the Pakistan National Shipping Corporation (PNSC) — Pakistan's national maritime enterprise — and reflects realistic patterns in seaborne trade across Pakistan's major ports (Karachi Port, Port Qasim, and Gwadar).
The dataset covers 5,000 shipment records spanning tanker voyages, bulk carrier trade, container shipping, and break-bulk cargo — providing a comprehensive view of Pakistan's import/export flows via sea.
📦 Data Source & Methodology
- Inspiration: Pakistan National Shipping Corporation (PNSC)
- Type: Synthetic (algorithmically generated using pandas & numpy)
- Realism: Exchange rates, freight rates, cargo types, port names, HS codes, and vessel specifications reflect real-world Pakistan maritime trade conditions
- Scope: 2020–2026 | Imports & Exports | All major cargo segments
🗂️ Feature Descriptions (47 Columns)
🔖 Identifiers & Time
| Column |
Type |
Description |
Shipment_ID |
String |
Unique identifier per shipment (e.g., PKT-2022-00123) |
Year |
Integer |
Year of departure (2020–2026) |
Month |
Integer |
Month of departure (1–12) |
Quarter |
String |
Fiscal quarter (Q1–Q4) |
📅 Dates
| Column |
Type |
Description |
Departure_Date |
Date (YYYY-MM-DD) |
Date vessel left the port of loading |
Arrival_Date |
Date (YYYY-MM-DD) |
Date vessel arrived at port of discharge |
Transit_Days |
Integer |
Sea transit duration in days |
Customs_Clearance_Days |
Float |
Days taken for customs clearance at destination |
🌍 Trade Information
| Column |
Type |
Description |
Trade_Direction |
Categorical |
Import or Export |
Trade_Category |
Categorical |
Sector: Energy, Agri-commodities, Manufactured Goods, etc. |
Cargo_Type |
Categorical |
Specific commodity (e.g., Crude Oil, Wheat, Coal, Textiles) |
HS_Code |
String |
Harmonized System customs classification code |
Incoterms |
Categorical |
Trade terms: CIF, FOB, CFR, EXW, DDP, FAS |
Payment_Terms |
Categorical |
LC, DP, DA, Open Account, Advance Payment |
Currency |
Categorical |
Transaction currency (USD, EUR, AED, PKR, etc.) |
🏗️ Ports
| Column |
Type |
Description |
Port_of_Loading |
String |
Origin port name and country |
Port_of_Discharge |
String |
Destination port name |
🚢 Vessel Information
| Column |
Type |
Description |
Vessel_Name |
String |
Name of the vessel |
Vessel_Type |
Categorical |
Vessel category (VLCC, Aframax, Panamax Bulk Carrier, etc.) |
Vessel_Flag |
Categorical |
Country of vessel registration |
Shipping_Line |
Categorical |
Operating carrier (PNSC, Maersk, CMA CGM, etc.) |
Vessel_DWT_MT |
Integer |
Deadweight tonnage capacity (metric tons) |
Gross_Tonnage_GT |
Integer |
Gross tonnage of the vessel |
Vessel_Age_Years |
Float |
Age of the vessel in years |
Average_Speed_Knots |
Float |
Average sailing speed during voyage |
📦 Cargo
| Column |
Type |
Description |
Cargo_Volume_MT |
Float |
Cargo weight in metric tons |
TEU_Count |
Float |
Container count (only for container ships; NaN otherwise) |
💰 Financials
| Column |
Type |
Description |
Freight_Rate_USD_per_Unit |
Float |
Rate per MT (or per TEU for containers) in USD |
Total_Freight_USD |
Float |
Total freight cost in USD |
Cargo_Value_USD |
Float |
Estimated CIF/FOB value of goods in USD |
PKR_USD_Exchange_Rate |
Float |
PKR/USD rate at time of shipment |
Cargo_Value_PKR |
Float |
Cargo value converted to Pakistani Rupees |
Port_Charges_USD |
Float |
Port dues, pilotage, berthing fees in USD |
Demurrage_Days |
Float |
Excess days in port beyond allowed laytime |
Demurrage_Cost_USD |
Float |
Financial penalty for excess port time |
Import_Duty_Pct |
Float |
Customs import duty percentage |
Import_Duty_USD |
Float |
Calculated import duty in USD |
Insurance_Pct |
Float |
Cargo insurance rate (% of cargo value) |
Insurance_Cost_USD |
Float |
Total cargo insurance cost in USD |
Bunker_Fuel_Consumed_MT |
Float |
Total bunker fuel consumed on voyage (MT) |
Bunker_Price_USD_per_MT |
Float |
Market price of bunker fuel per MT |
Bunker_Cost_USD |
Float |
Total fuel cost in USD |
Total_Voyage_Cost_USD |
Float |
Sum of freight + port + demurrage + bunker + insurance |
✅ Customs & Compliance
| Column |
Type |
Description |
Customs_Cleared |
Binary (Yes/No) |
Whether customs clearance was successfully obtained |
ISM_Certified |
Binary (Yes/No) |
International Safety Management certification status |
Safety_Incident |
Categorical |
Incident severity: None, Minor, Moderate |
Environmental_Flag |
Categorical |
Environmental compliance: Clean, Minor Violation |
🤖 Potential Machine Learning Tasks
This dataset is well-suited for a variety of ML tasks:
1. 📈 Regression
- Freight Rate Prediction: Predict
Freight_Rate_USD_per_Unit based on vessel type, cargo type, route, and season
- Total Voyage Cost Estimation: Predict
Total_Voyage_Cost_USD using vessel specs, route, cargo volume, and market conditions
- Demurrage Cost Forecasting: Predict
Demurrage_Cost_USD from cargo type, vessel type, and port
- Cargo Value Estimation: Predict
Cargo_Value_USD from cargo type, volume, and trade direction
2. 🏷️ Classification
- Customs Clearance Prediction: Predict whether a shipment will be cleared (
Customs_Cleared = Yes/No)
- Safety Incident Classification: Classify shipments by risk level (
Safety_Incident: None/Minor/Moderate)
- Environmental Compliance: Flag potential violations (
Environmental_Flag)
- Trade Direction Classification: Classify shipments as Import or Export from voyage characteristics
3. ⏱️ Time Series Analysis
- Monthly Trade Volume Trends: Forecast Pakistan's seaborne trade volumes over time
- Freight Rate Seasonality: Detect seasonal patterns in shipping costs
- PKR Exchange Rate Impact Analysis: Study correlation between PKR/USD rate and trade volumes
4. 🔍 Clustering / Unsupervised Learning
- Shipment Profiling: Cluster shipments by cost structure and route efficiency
- Port Congestion Analysis: Group shipments by clearance delays and demurrage
- Vessel Performance Segmentation: Cluster vessels by fuel efficiency and speed
5. 🔎 Anomaly Detection
- Fraud / Undervaluation Detection: Detect shipments with unusually low declared cargo values
- Outlier Voyage Cost Detection: Identify abnormally expensive or cheap voyages
⚠️ Disclaimer
This is a synthetic dataset generated for educational and research purposes. All shipment IDs, vessel names, and financial figures are algorithmically simulated. They do not represent actual PNSC transactions or real trade data. The dataset is intended for ML model development, data science practice, and exploratory analysis.
