Zuber Ride Analysis

Table of Contents

• Zuber Ride Analysis
  • Table of Contents
  • Project Objective
  • Project Structure
  • Tools and Techniques Utilized:
  • Specific Results and Outcomes
  • What I Have Learned from This Project
  • How to Use This Repository

Project Objective

This project aims to analyze and identify patterns in the available data for Zuber, a ride-sharing company launching in Chicago. The goal is to understand passenger preferences and the impact of external factors, such as weather, on rides. The project involves working with a database of taxi rides and weather data to test hypotheses and extract actionable insights.

back to top

Project Structure

1. Data Exploration and Import

• Objective: Understand the data schema and import the necessary files.
• Actions Taken:
  • Reviewed the schema of the taxi rides and weather data, which included tables for neighborhoods, cabs, trips, and weather_records.
  • Imported CSV files containing key data: project_sql_result_01.csv for taxi companies and rides, project_sql_result_04.csv for average trips by neighborhood, and project_sql_result_07.csv for trip duration and weather conditions.
• Outcome: Ensured that the data was correctly imported and explored for any potential inconsistencies or areas requiring cleaning.

2. Data Cleaning and Preparation

• Objective: Ensure the dataset is ready for analysis by handling missing values, incorrect data types, and other issues.
• Actions Taken:
  • Checked each column for missing values and corrected or filled in these values where necessary.
  • Converted columns to their correct data types (e.g., timestamps, numeric fields).
  • Merged tables where necessary, particularly trips and weather_records, based on the start time of the trip and weather records.
• Outcome: A clean, merged dataset was prepared for further analysis and visualization.

3. Exploratory Data Analysis (EDA)

• Objective: Analyze trends, identify patterns, and visualize the data to gain insights into ride behavior.
• Actions Taken:

  • Company Analysis:

    • Counted the number of taxi rides for each company over specific dates to understand their market share.
    • Aggregated ride counts for companies with "Yellow" or "Blue" in their names for a comparative analysis.
    • Identified the most popular taxi companies for November 2017, such as Flash Cab and Taxi Affiliation Services, and grouped all other companies as "Other."

  • Neighborhood Analysis:

    • Analyzed the top 10 neighborhoods in Chicago where rides ended, based on average trips.
    • Visualized the drop-off locations and compared the distribution of trips across neighborhoods.
• Outcome: Provided insights into which taxi companies had the most rides and which neighborhoods were the most popular drop-off points, laying the groundwork for further hypothesis testing.

4. Weather and Ride Duration Analysis

• Objective: Investigate the relationship between weather conditions and ride duration.

• Actions Taken:

  • Classify Weather Conditions:
    • Retrieved weather records for each hour, categorizing weather as "Bad" if the description contained "rain" or "storm," and as "Good" otherwise.
  • Trip Duration and Weather Analysis:
    • Filtered the dataset to analyze only rides starting in the Loop neighborhood and ending at O'Hare Airport.
    • Linked these trips with weather conditions to see how rainy weather affected ride durations on Saturdays.

• Outcome: The merged data allowed for a deeper understanding of how external factors like weather could impact the ride-sharing experience.

5. Visualizations

• Objective: Create meaningful visualizations to represent data findings.
• Actions Taken:
  • Plotted graphs showing the number of rides per company to identify market leaders and compare their performance.
  • Created bar charts and histograms to display the distribution of rides across neighborhoods.
  • Developed scatter plots and line graphs to examine the impact of weather conditions on ride durations.
• Outcome: Visualizations helped communicate the data trends clearly and supported the decision-making process.

6. Hypothesis Testing

• Objective: Test whether weather conditions significantly impact the duration of taxi rides.
• Actions Taken:
  • Hypothesis Testing for Ride Duration:
  • Formulated null and alternative hypotheses to test whether the average duration of rides from the Loop to O'Hare Airport changes on rainy Saturdays.
  • Used appropriate statistical tests (e.g., t-test or Mann-Whitney U test) based on the distribution and normality of the data.
  • Set an alpha level (commonly 0.05) to determine the significance of the test results.
• Outcome: The hypothesis test revealed the extent to which weather conditions affect ride durations, providing insights into ride-sharing behavior during different weather scenarios.

back to top

Tools and Techniques Utilized:

• SQL: To query and analyze taxi and weather data, identifying ride patterns and correlations.
• Pandas and NumPy: For data manipulation, cleaning, and transformations.
• Matplotlib and Seaborn: For visualizing data trends, distributions, and relationships.
• Scipy: For statistical analysis and hypothesis testing.
• Beautiful Soup: Utilized for web scraping to retrieve weather data for Chicago in November 2017 from HTML pages. This allowed for further analysis on the impact of weather conditions on ride patterns.

back to top

Specific Results and Outcomes

Taxi Company Analysis

• Top Companies for Rides:

  • Flash Cab and Taxi Affiliation Services were found to be the most popular taxi companies in Chicago during November 2017.
  • Flash Cab had the highest number of trips over a two-day period, with a significant gap compared to other companies.

• Companies with "Yellow" or "Blue" in Their Name:

  • Analysis focused on companies with "Yellow" and "Blue" in their names, revealing that these companies accounted for a substantial portion of rides between November 1-7, 2017.
  • This provided insight into branding and ride preferences, showing that taxis branded with colors may have different customer perceptions or market niches.

• Market Share:

  • When grouping all companies except the top two (Flash Cab and Taxi Affiliation Services) as "Other," Flash Cab still led in market share, indicating a possible monopoly or strong brand preference in the taxi market.

Neighborhood Ride Patterns

• Top Drop-Off Neighborhoods:

  • Loop, River North, and Streeterville were among the top neighborhoods for ride drop-offs, making them key areas of demand in Chicago.
  • Loop had the highest average drop-offs in November 2017, suggesting that it is a central hub for taxi services, likely due to its commercial and business activity.

• Visualization of Demand:

  • The top 10 neighborhoods showed a clear trend: central areas with business and entertainment activities received more rides compared to residential or outlying neighborhoods.

• Average Trips per Neighborhood:

  • A chart of the average number of trips highlighted that certain neighborhoods consistently receive a high volume of drop-offs, indicating consistent demand throughout November.

Analysis of Weather Impact on Ride Duration

• Correlation Between Weather and Rides:
  • Bad weather conditions (e.g., rain or storms) showed a slight increase in ride duration for trips from the Loop to O'Hare International Airport.
• Ride Duration on Saturdays:
  • For Saturdays, ride duration showed some variability when comparing rainy vs. non-rainy conditions, indicating that weather could affect traffic and road conditions, thus extending trip times.
• Hypothesis Testing:
  • The hypothesis test revealed that ride durations were, in fact, longer on rainy Saturdays, indicating that weather does have a statistically significant impact on ride duration. However, the effect size was not very large, implying that while weather impacts ride times, it may not be the only contributing factor.

Rides by Taxi Company and Neighborhood

• Flash Cab Dominance:

  • Flash Cab led in the number of rides over a selected two-day period in November, followed closely by Taxi Affiliation Services. Other companies had significantly fewer trips, suggesting strong competition between these two major players.

• Neighborhood Analysis & Popularity:

  • Central business and tourist neighborhoods such as Loop, Near North Side, and River North received the highest number of trips, making them focal points for Zuber to target high demand.

• Consistency of Ride Demand:

  • Most neighborhoods with high drop-off rates had consistent demand throughout November, with very little fluctuation, suggesting stable market demand in these areas.

Duration Analysis from Loop to O'Hare

• Trips Duration Consistency:

  • For trips from Loop to O'Hare, average ride durations varied depending on the time of day and weather conditions.
  • On rainy days, rides took slightly longer, which is crucial for estimating travel time for passengers traveling to the airport.

• Analysis of Duration Trends:

  • The average duration of rides from Loop to O'Hare was around 45-55 minutes on non-rainy Saturdays, while on rainy Saturdays, the duration increased to around 55-65 minutes.

Data Visualization Insights

• Ride Count by Taxi Company:
  • Bar charts revealed the dominance of Flash Cab and Taxi Affiliation Services over other smaller companies.
• Neighborhood Drop-off Visualization:
  • Bar charts of the top 10 neighborhoods showed clear hotspots of activity where Zuber could concentrate its services.
• Impact of Weather on Ride Duration:
  • Visualizations like boxplots and scatter plots highlighted the difference in ride duration based on weather conditions, clearly showing the impact of rainy weather on extending ride times.

back to top

Key Recommendations for Zuber

• Focus on High-Demand Neighborhoods: Target high-demand neighborhoods like Loop, Near North Side, and River North for ride-sharing services, as they are popular drop-off destinations.
• Prepare for Weather Variations: Since ride durations are affected by weather, Zuber can develop strategies like surge pricing, optimized routing, and alerts to customers for better service during adverse weather conditions. -Compete with Market Leaders: Understand the strong market presence of Flash Cab and Taxi Affiliation Services. Zuber should aim to compete on price, service quality, or availability to gain market share in Chicago.
• Monitor Ride Duration to O'Hare: Since O'Hare is a popular destination, Zuber can ensure accurate estimated times for airport trips and develop marketing strategies to target passengers traveling from central neighborhoods to the airport.

back to top

What I Have Learned from This Project

• Data Manipulation and Cleaning: Gained experience in handling real-world data, including dealing with missing values, data type conversions, and rounding numerical values for analysis.
• SQL Proficiency: Improved ability to query large datasets, filter data based on conditions, and perform joins across multiple tables to derive insights.
• Hypothesis Testing: Developed skills in forming null and alternative hypotheses, determining appropriate significance levels (alpha), and using statistical methods to test hypotheses.
• Data Visualization: Enhanced skills in creating meaningful visualizations that illustrate trends and comparisons across different datasets (e.g., taxi company rides and neighborhood drop-off counts).
• Exploratory Data Analysis (EDA): Applied EDA techniques to uncover patterns in data, understand distributions, and make data-driven decisions.

back to top

How to Use This Repository

1. Clone the Repository
   git clone https://github.com/realdanizilla/Zuber.git

2. Review Jupyter Notebooks and Datasets
   Explore the datasets, step-by-step data analysis, preprocessing, and hypothesis testing related to taxi rides and weather.

3. Execute Notebooks for Analysis
   Run the notebooks to understand the data exploration process, visualizations, and hypothesis testing performed.

4. Examine Results and Insights
   Analyze conclusions drawn from taxi trip data to optimize business strategies.

back to top