Project Overview

This project implements an automated, end-to-end "Transit & Weather Statistic" data pipeline focused on the Turku region. The analytical core of this project tests the hypothesis that adverse weather conditions (meaning heavy precipitation, freezing temperatures) can cause a quantifiable shift in human behavior. Specifically, a decrease in public transit usage and a concurrent increase in personal vehicle traffic, which manifests as an increased volume of parking ticket violations.

Architecture & Tech Stack

The system is designed with a strict separation of concerns, utilizing a modern data lakehouse system alongside a compiled microservice backend and a reactive frontend dashboard.

• Data Engineering & ETL: Databricks, PySpark
• Machine Learning: Apache Spark MLlib
• Backend API: Go (Golang)
• Frontend: Vue.js, Nuxt, NuxtUI

Data Sources (2023)

• Weather Data (FMI): Daily precipitation and temperature statistics collected from the Turku Artukainen measuring station.
• Public Transit Data (Föli): Daily passenger boarding counts registered via ticketing machines across the Turku city limits.
• Parking Fine Data: Records of parking violations and tickets issued within the Turku area.