I build end-to-end data pipelines — from raw public datasets to interactive dashboards.
Focused on logistics & transport analytics, SQL, Python ETL, and Power BI.
SQL + Power BI end-to-end analytics for a fictional logistics company
Stack: PostgreSQL · Power BI · DAX
What I built:
- Relational schema with 5 tables (
shipments,carriers,routes,costs,delays) - 1 000-row synthetic dataset with intentional anomalies for data quality demo
- Data cleansing pipeline in SQL: duplicate removal, NULL imputation, outlier reclassification
- 3 analytical views:
vw_cost_per_km,vw_on_time_rate,vw_monthly_trends - Power BI dashboard: cost overview + delivery performance pages, built-in forecast
Key KPIs: Cost per km · On-time delivery rate · Delay rate · Monthly cost trend
Real Eurostat data (road_go_ta_tott) analysed with Python → PostgreSQL → Power BI
Stack: Python (pandas) · PostgreSQL · SQLAlchemy · Power BI · DAX
What I built:
- Eurostat API client parsing JSON-stat 2.0 → flat CSV
- Multi-step ETL: missing value strategy, country code standardisation (ISO 3166-1), unit normalisation
- Star schema:
fact_freight+ 3 dimension tables (dim_country,dim_year,dim_cargo_type) - 5 analytical SQL views including YoY growth with LAG window functions and market share %
- Power BI dashboard: Europe filled map + country rankings + YoY trend with cross-filtering
Key KPIs: Total volume (tonnes) · YoY Growth % · Market Share % · CAGR