Uber Ride Patterns in Extreme Weather

Project Description

The aim of this project is to analyze the relationship between weather conditions and Uber ride usage, especially during extreme weather conditions. The problem we address is whether weather conditions, ranging from subzero temperatures to scorching heat, influence the frequency of ride-sharing usage.

• Compare the Uber rides vs the weather in New York City
• Compare the following data:
  • Ride demand
  • Weather conditions
  • Geographic distribution

Dataset

We used the following data for our analysis:

• Historical weather data
  • Dates: January 2014 - December 2015
  • Location: New York City
  • Units: Imperial
  • Weather data includes: Temperature, Max & Min Temperature, Feels Like, Humidity, Clouds, Pressure, Conditions, Rain, Snow, Dew Point, Visibility, Wind
• Uber pickups in New York City

Research Questions

1. Correlation Study: Is there a specific correlation between weather factors (e.g., temperature, precipitation) and the number of rides?
2. Demand Impact: How does weather influence ride demand for Uber and other FHV companies?
3. Strategic Planning: Can correlations help plan promotions and increase ride shares?

Hypothesis

Hypothesis: Extreme weather conditions significantly affect the demand and geographic distribution of rides.

Null Hypothesis: Extreme weather conditions do not significantly impact ride demand or distribution.

Data Analysis

Interpretation & Recommendations

The strongest observed relationship was between temperature and number of trips—higher temperatures corresponded with fewer trips. Humidity and cloudiness had weaker, yet present, correlations.

Recommendations:

• Promote Services on Colder Days: Offer discounts to encourage more usage during cold weather.
• Optimize Fleet Deployment: Reduce idle time by adjusting driver availability on hot days.
• Explore Humidity and Cloudiness Effects: These may reveal additional insights with more detailed thresholds.

Project Limitations

• Relied on static datasets due to lack of accessible APIs.
• Data gaps (April–September 2014 missing; only partial 2015 data).
• Geographical scope limited to NYC (not all boroughs equally represented).
• Time constraints due to class schedule and external commitments.

Conclusion

Our analysis supports the hypothesis that extreme weather conditions significantly influence the demand for Uber and other FHV (For-Hire Vehicle) services in New York City. We observed a decrease in ride volume as temperatures increased, suggesting alternative transportation use or reduced travel during hotter weather.

These insights highlight business opportunities—ride-sharing companies can better align promotions and driver deployments with weather conditions to optimize engagement and efficiency.

However, the project was limited by dataset scope and availability. Future studies with broader and more complete data could provide even deeper insights into weather’s role in urban mobility.

In summary, weather is a meaningful factor in ride-share demand. Recognizing and responding to this can empower smarter urban transportation strategies.