IoT-Based Smart Parking System Dataset - 2-Year Occupancy Data

Dataset Overview

This IoT Smart Parking dataset published in January 2024 provides an exceptionally long monitoring period of 2 complete years of parking facility operations. Collected through ThingSpeak IoT platform using IR sensors and ESP32 microcontrollers, it offers unprecedented insights into urban parking dynamics.

IoT System Architecture

Hardware Components

• IR Proximity Sensors: Infrared sensors detecting vehicle presence in each parking slot
• ESP32 Microcontroller: WiFi-enabled board processing sensor data and cloud transmission
• Power Supply: Reliable power infrastructure ensuring continuous 24/7 operation
• Sensor Network: Multiple sensors covering entire parking facility

Cloud Platform

• ThingSpeak Integration: IoT analytics platform for real-time data logging and visualization
• MQTT Protocol: Lightweight messaging for efficient sensor-to-cloud communication
• API Access: RESTful APIs for data retrieval and third-party integration

Data Features

Parking Occupancy Metrics

• Slot Status: Binary occupied/vacant status for each individual parking space
• Total Occupancy: Number of occupied slots at each timestamp
• Availability Rate (%): Percentage of vacant spaces
• Timestamps: Precise date-time for every status change

Temporal Patterns

• Hourly Distribution: Peak usage hours (morning rush, lunch, evening)
• Daily Patterns: Weekday vs weekend variations
• Seasonal Trends: Changes across months and seasons over 2 years
• Special Events: Anomalous patterns during holidays or local events

Parking Duration

• Stay Time: How long vehicles occupy slots (short-term vs long-term)
• Turnover Rate: Frequency of slot usage per day
• Utilization Efficiency: Metrics for parking facility performance

Urban Transportation Research Applications

Parking Availability Prediction

• Train ML models forecasting parking availability 15-60 minutes ahead
• Enable mobile apps guiding drivers to available spaces
• Reduce cruising time and associated emissions

Dynamic Pricing Optimization

• Implement demand-based pricing during peak hours
• Balance occupancy across time periods
• Maximize revenue while ensuring availability

Urban Planning

• Determine optimal parking facility sizing based on actual demand patterns
• Identify over-supplied or under-supplied areas
• Support zoning and development decisions

Traffic Management

• Correlate parking occupancy with nearby traffic congestion
• Design coordinated parking and traffic flow systems
• Reduce circulating traffic from parking search

Machine Learning Tasks

• Time-Series Forecasting: Predict future occupancy using ARIMA, LSTM, or Prophet models
• Classification: Categorize time periods as low/medium/high demand
• Anomaly Detection: Identify unusual patterns indicating sensor failures or special events
• Regression: Model relationships between time/weather/events and parking demand

Smart City Integration

The dataset demonstrates practical IoT deployment for smart city services. The 2-year duration provides robustness against seasonal anomalies and enables validation of long-term prediction models. ThingSpeak platform integration shows scalable cloud-based IoT architecture applicable to other smart city sensors.

Data Quality Advantages

• Long Duration: 2 years ensures statistical significance and captures rare events
• High Frequency: Near real-time updates enabling responsive applications
• Completeness: Minimal gaps from reliable ESP32/ThingSpeak infrastructure
• Validated: IR sensors provide accurate binary occupancy detection