Dark Mode
DC Bike Share Demand Forecasting Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This product features daily bike sharing demand data paired with corresponding local weather conditions for the Washington DC metro area. It covers customer counts (both registered and casual users) alongside detailed meteorological information, including temperature variations, precipitation levels, wind speed, and various specific weather event indicators (e.g., fog, thunder, snow). The data is essential for developing predictive models to forecast future bike usage, a critical task for shared mobility operators aiming to manage variable demand and address logistical challenges like equipment redistribution and potential shortages or oversupply. Given the role of shared mobility in mitigating climate change, understanding demand patterns is vital for sustainable urban planning.
Columns
The dataset contains 29 columns detailing time, temperature, weather events, and customer counts.
- date: The calendar date in YYYY-MM-DD format.
- temp_avg: The average daily temperature in degrees Celsius.
- temp_min: The minimum daily temperature in degrees Celsius.
- temp_max: The maximum daily temperature in degrees Celsius.
- temp_observ: The temperature measured at the specific time of observation in degrees Celsius.
- precip: The total amount of daily precipitation in millimeters.
- wind: The average daily wind speed in meters per second.
- wt_fog: Indicator for fog, ice fog, or freezing fog (may include heavy fog).
- wt_heavy_fog: Indicator for heavy fog or heavy freezing fog.
- wt_thunder: Indicator for thunder.
- wt_sleet: Indicator for ice pellets, sleet, snow pellets, or small hail.
- wt_hail: Indicator for hail.
- wt_glaze: Indicator for glaze or rime.
- wt_haze: Indicator for smoke or haze.
- wt_drift_snow: Indicator for blowing or drifting snow.
- wt_high_wind: Indicator for high or damaging winds.
- wt_mist: Indicator for mist.
- wt_drizzle: Indicator for drizzle.
- wt_rain: Indicator for rain (may include freezing rain, drizzle, and freezing drizzle).
- wt_freeze_rain: Indicator for freezing rain.
- wt_snow: Indicator for snow, snow pellets, snow grains, or ice crystals.
- wt_ground_fog: Indicator for ground fog.
- wt_ice_fog: Indicator for ice fog or freezing fog.
- wt_freeze_drizzle: Indicator for freezing drizzle.
- wt_unknown: Indicator for unknown source of precipitation.
- casual: The daily count of unregistered customers.
- registered: The daily count of registered customers.
- total_cust: The sum of registered and casual customers for the day.
- holiday: A binary indicator showing whether the day is a public holiday.
Distribution
This data product contains daily records spanning an eight-year period. The primary file is typically provided in CSV format. There are approximately 2,922 daily records included, starting from 1 January 2011 through 31 December 2018. Note that certain fields, such as average temperature, have a substantial number of missing values (around 28%), while the customer count fields have minimal missing data (less than 1%).
Usage
This data product is ideal for several applications in urban logistics and data science:
- Demand Forecasting: Predict daily, weekly, or monthly future bike demand using time series analysis techniques.
- Operational Planning: Optimise the redistribution of bikes between stations to prevent oversupply or shortages.
- Feature Importance: Identify which meteorological or calendar variables are most important for predicting bike usage.
- Business Intelligence: Determine appropriate pricing strategies or identify busy stations for advertisement placement.
- Pattern Analysis: Employ anomaly detection to uncover inherent seasonality and usage trends in customer data.
Coverage
The geographic scope covers the Washington DC metro area. The time period runs from the beginning of 2011 to the end of 2018. This dataset focuses specifically on daily demand and general regional weather conditions. Determining new station locations, analysing movement patterns, or planning specific routes requires pairing this data with additional geo-spatial information.
License
CC BY-NC-SA 4.0
Who Can Use It
- City Planners: To evaluate the effectiveness and expansion potential of shared mobility systems.
- Bike Sharing Operators: To improve fleet management, maintenance scheduling (heavy bike use leads to more breakdowns), and resource allocation.
- Data Scientists/Analysts: To build and validate predictive models for transport demand.
Dataset Name Suggestions
- DC Bike Demand and Climate Records (2011-2018)
- Washington Metro Daily Bike Sharing Usage
- DC Bike Share Demand Forecasting Data
- Weather Impact on Shared Mobility Demand
Attributes
Original Data Source: DC Bike Share Demand Forecasting Data
QUALITYLoading...