Shared bicycle data

Comprehensive Spatiotemporal Dataset of Shared Bicycle Operations in Xiamen, China IMPORTANT: Data Usage Agreement BY DOWNLOADING THIS DATASET, YOU AGREE TO THE FOLLOWING TERMS: 1. Non-Commercial Use Only This dataset is provided for **academic research and educational purposes ONLY**. Commercial use is **strictly PROHIBITED** and violates the original data usage agreement. 2. Required Attribution You **MUST** cite: - Original data providers:** Xiamen shared bicycle operators via the Digital China Innovation Competition 3. Privacy Protection Requirements You **MUST NOT**: - Attempt to identify individuals, households, or specific residential addresses - Reverse-engineer user identities or personal travel patterns - Conduct individual-level identification or tracking - Use the data in ways that could compromise user privacy 4. Aggregated Analysis Only All analyses **MUST** be conducted at aggregated levels (neighborhood, district, or city-wide). Individual-level analysis is **PROHIBITED**. 5. No Redistribution Under Different Terms You may share this dataset, but **ONLY** under the same CC BY-NC 4.0 license and usage terms. 6. Compliance Your use must comply with: - Research ethics standards - Data protection regulations - The original competition data usage agreement Dataset Overview Temporal Coverage: December 21-25, 2020 (Monday-Friday), 6:00-10:00 a.m. (UTC+8)Spatial Coverage: Urban area of Xiamen, Fujian Province, ChinaFleet Size: 53,630 unique bicyclesCoordinate System: WGS-84 Dataset Components 1. Order Data (220,675 validated trip records) Origin-destination pairs with precise GPS coordinates (6 decimal places) Lock status changes (unlock/lock events) High-resolution timestamps Trip duration information 2. Trajectory Data (2,849,243 GPS tracking points) Continuous GPS tracking during trips Second-level positioning accuracy Complete movement paths between origins and destinations 3. Electronic Fence Data (14,071 authorized parking zones) Unique fence identifiers and names Four-corner coordinates defining rectangular parking areas Polygonal boundary definitions 4. Infrastructure Data Road network information (OSM-derived bicycle-accessible routes) Hourly weather observations (temperature, precipitation, wind, humidity, pressure) 5. Processing Scripts Python data cleaning algorithms MATLAB clustering and prediction models File Structure Shared bicycle data/ ├─ Algorithm data/ # MATLAB scripts for clustering & prediction ├─ Data cleaning algorithm/ # Python scripts for preprocessing ├─ Electronic fence data/ # Fence IDs, names, and coordinates ├─ Order data/ # Trip orders and vehicle state logs ├─ Trajectory data/ # GPS traces by bicycle and time └─ Weather data/ # Hourly weather observations Data Format and Quality Format: CSV (Comma-Separated Values) with UTF-8 encodingTime Zone: China Standard Time (UTC+8) Units: Distance: meters (m) Duration: seconds Temperature: Celsius (°C) Wind speed: meters per second (m/s) Precipitation: millimeters (mm) Pressure: hPa Quality Assurance: Rigorous preprocessing with temporal filtering (60-3,600 seconds) Completeness verification and anomaly removal Cross-day correlation validation (r > 0.90) Cryptographic hashing of all bicycle IDs for privacy protection Key Variables Order Data: BICYCLE_ID: Unique bicycle identifier LATITUDE, LONGITUDE: GPS coordinates (WGS-84) LOCK_STATUS: 0 (unlocked/trip start) or 1 (locked/trip end) UPDATE_TIME1, UPDATE_TIME2: Trip start and end timestamps ORDER_DURATION: Trip duration in seconds Trajectory Data: BICYCLE_ID: Bicycle identifier LOCATING_TIME: GPS positioning timestamp LATITUDE, LONGITUDE: Real-time coordinates Electronic Fence Data: FENCE_ID: Unique fence identifier FENCE_NAME: Parking zone name LAT1-LAT4, LNG1-LNG4: Four-corner coordinates Weather Data: TIME: Observation timestamp TEMP: Air temperature RHUM: Relative humidity PRCP: Precipitation WSPD: Wind speed PRES: Atmospheric pressure COCO: Weather condition code