Skip to main content

Get Started with Dremio Cloud

To get started, sign up for an account at dremio.com/get-started and follow the guided setup to create your first project.

This guide shows you how to analyze transportation data and find usage patterns using natural language queries with Dremio's AI Agent. You'll work with the dremio_samples.nyc_citibikes.citibikes table that contains over 115 million bike-sharing records from New York City—real patterns from one of the world's busiest transportation networks.

Step 1: Discover and Explore Your Data​

Understanding your dataset structure is crucial before diving into analysis. Start by getting an overview of the bike-sharing data to understand what insights are possible.

On the homepage, enter the following prompt into Dremio's AI Agent chat box: Give me an overview of the nyc_citibikes.citibikes dataset.

The AI Agent uses capabilities like semantic search to find relevant datasets. Review the response to understand the schema, field definitions, and what types of analysis are possible with bike-sharing data.

Step 2: Compare User Behavior​

Ask a question in natural language to compare how different user types interact with the bike-sharing service. This analysis reveals usage patterns that inform operational decisions.

Ask the AI Agent to compare how subscribers and casual riders use the service: Give me the total number of rides and the average trip duration grouped by user type across the dataset.

The AI Agent generates and runs SQL using Dremio's query engine to answer this question. The results show that subscribers typically show higher ride frequency but shorter durations, indicating commuter behavior. Casual riders often have longer trips but lower frequency, suggesting leisure or tourist usage patterns. This analysis reveals how different user types require different operational strategies for bike availability, pricing models, and infrastructure investment.

Step 3: Analyze Peak Demand Patterns​

Dive deeper to understand when different user groups are most active. This temporal analysis provides insights for operational planning and resource allocation.

Ask the AI Agent to reveal hourly demand patterns: Analyze hourly ride patterns to find when demand peaks. Show a line chart of total rides per hour of the day, separated by user type, and include a short report highlighting the busiest hours for each group and recommendations for bike availability.

The AI Agent will create a clear chart showing distinct patterns. Subscribers peak during rush hours (8-9 AM, 5-6 PM) indicating commuter usage, while casual riders peak during midday and weekends showing leisure patterns. This temporal analysis reveals opportunities for dynamic pricing during peak hours, optimal timing for maintenance during low-demand periods, and capacity planning insights for fleet optimization.

Step 4: Run Comparative Analysis​

Use the AI Agent to identify the most influential factors affecting rider behavior. This analysis compares multiple variables to determine primary drivers of ridership patterns.

Ask the AI Agent to run a comprehensive comparative analysis: Run comparative analysis on seasonal, daily, hourly, and bike type patterns to identify which factor has the most significant impact on ride behavior. Then create a detailed visualization of the most influential factor.

The AI Agent will compare multiple variables and automatically identify which factor has the biggest impact on rider behavior, complete with actionable recommendations. This analysis reveals primary drivers of ridership, correlation insights between variables, and predictive indicators for forecasting usage patterns.

Step 5: Try Your Own Analysis​

Now that you understand how to analyze transportation data with the AI Agent, try exploring other bike-sharing questions.

You can also analyze your own data using Dremio's AI Agent.

Summary​

You have completed a transportation data analysis using natural language. You explored the dataset structure, compared user behavior patterns, analyzed peak demand times, and identified influential factors affecting ridership. Discovery, exploration, and analysis that could previously take hours can now be done in minutes with Dremio's AI Agent.

Troubleshoot​

  • If a prompt doesn't work as expected, try simplifying the request or verifying that you're referencing the correct dataset (nyc_citibikes.citibikes).
  • Contact your administrator if sample data isn't available in your environment.