mcken-04/apex-fleet-data-pipeline

This project is a portfolio style data pipeline built around a made up logistics company called Apex Regional Deliveries. The imagined company was struggling with unreliable data about its delivery drivers, its fuel usage, and its vehicle maintenance records, and this project shows one way to fix that by building a full pipeline from raw data all the way to a finished business dashboard. It starts by generating over ten thousand rows of realistic but fake logistics data using Python, covering drivers, vehicles, and delivery routes. The data is deliberately made messy on purpose, with problems like negative fuel amounts and missing maintenance records mixed in, so the project can demonstrate how to catch and handle those kinds of real world data quality issues rather than working with a clean, perfect dataset. That raw data is then loaded into a PostgreSQL database, where SQL scripts clean it up, calculate useful values, and flag anomalies using views built directly in the database. From there, the cleaned data feeds into Power BI, where it is organized into a proper data model and turned into an interactive dashboard with custom calculations and visual filters. The finished dashboard is designed to answer specific business questions, such as which drivers are consistently slower than others, which vehicles have fuel readings that are mathematically impossible and likely point to a sensor or data entry problem, and which vehicles are missing or overdue for maintenance and therefore a safety or compliance risk. The project also includes written business recommendations based on what the dashboard reveals, like auditing fuel sensors and pulling overdue vehicles from service. This is best understood as a learning and portfolio project demonstrating an end to end analytics workflow, from generating and cleaning data through to building a decision ready business intelligence dashboard, and it would be most useful to someone studying or showcasing skills in data engineering, SQL, and Power BI reporting.