NYC Yellow Taxi Data Usage and Profitability

Posted on Oct 18, 2019
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

NYC Yellow Taxi Data Usage and Profitability

Purpose of the Data Study

The study’s main objective is to understand some of the characteristics of the yellow cab industry and how some data variables such as distance, speed or traffic affect profitability.

 

Overview

During 2019, the NYC Yellow Taxi industry will have revenue of about $1.6 billion ($4.5 million per day) and will transport more than 140 million passengers (~400 k/day). Profits are estimated to be around $300 million (~$900k/day).The total amount of distance traveled (with and without passengers) in a day is ~1.2 million miles. To put distance in perspective, it would take a single person about 20 years to travel it.

NYC Yellow Taxi Data Usage and Profitability

Data on Characteristics of Yellow Taxi Usage

We can characterize the usage of NYC taxi as a small distance and mid fare service. About 80% of the trips have a distance of 5 miles or less and a fare of $18 or less.

For example, trips with longer distances such as Manhattan to and from JFK Airport only represent a small percentage of the total.

NYC Yellow Taxi Data Usage and Profitability
NYC Yellow Taxi Data Usage and Profitability

Fig.Β 1Β Β Concentration of trips by Distance and Fare (random sample)

Purpose of Trips

Even though we don’t have much information regarding the purpose of the trips, we can make some assumptions by looking at the most traveled routes and cross-check it with the hour. In the following heatmap we can identify how the demand is concentrated in particular hours of the day.

Fig.Β 2Β Heatmap of trips by day and hour

Takeaway 1: People use taxis when tired, especially after a long day at work

From the previous graph we can see that most of the trips are generated between 4 and 8pm. Even though there is slighter more demand from Thursday to Saturday, we see that in general the demand is still high at that particular time of the day any day of the week.

Now, by looking at the routes with more trips, we can tell that much of the traffic is generated within the Upper East Side.Β 

Typical Fare Composition vs Operating Costs

On average, a NYC trip costs the users $17.30 of which $14 goes directly to pay for the direct costs. At the end, the taxi’s profit per trip is under $4.

Fig.Β 3Β Average Revenue and Profit per trip

As the two main factors that affect cost depend on time or distance, the costs can be classified according to how they are incurred in the following way:

  • Time related: costs that fluctuate with distance
    • Variable: labor (driver’s salary)
  • Distance related: Costs that depend directly by miles traveled
    • Variable: gas and maintenance (tires, oil, breaks and other preventive maintenance)
    • Fixed and Semifixed: car depreciation, replacement car cost and insurance.

Sensibility Analysis: if we increase each of the main costs independently (keeping the rest of the variables constant) by 10% we can identify how sensible is each cost in the total profit.

Cost Variable Increase Profit/trip Cost Increase Profit Β 
Insurance 10% Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 3.73Β  Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 0.02Β  27% From 3,500 to $$3,850/yr
Gas Price 10% Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 3.67Β  Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 0.09Β  26% From 3.5 to $3.85/gallon
Car Cost 10% Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 3.64Β  Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 0.12Β  26% From 20k to $27.5k per car
Labor 10% Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 2.98Β  Β $Β Β Β Β Β Β Β Β Β Β Β Β Β Β 0.78Β  21% From 20 to $22/hr

From the previous table we can see that labor cost is one of the main factors that influences profit of the taxi business. To show this with more detail, lets look at the following example:

Β Β Β Β Β Β Β Β Β Β Β Β Origin: Upper East Side SouthΒ Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Destination: JFK Airport

Β Β Β Β Β Β Β Β Β Β Β Β Distance: 18 milesΒ Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Fare: $52

Β  Time of Day Minutes Cost Profit Profit %
Average Time All Hours / Add Days 49 Β $ 42Β  Β $ 10Β  19%
Fastest Thursday 12pm-4am 26 Β $ 29Β  Β $ 23 44%
Slowest Friday between pm4-8pm 70 Β $ 54Β  Β ($ 2) -3%

Takeaway 2: Speed is the most important factor in determining profit because costs associated with time (such as labor) have the biggest impact on cost

As seen in the previous example, just the decreased speed that is caused by heavier traffic (early morning vs peak hour) can render a profitable route unproductive. It is because basically the drivers lose so much time in moving so slowly that they miss out on picking up more fares (generating more revenue).

Conclusions

So with all this information, can we determine which is the best shift or the best time to work as a taxi driver? It is not that easy. Since drivers are not allowed to reject customers based on their destination, it is very difficult for the driver to control where they have to go.

It is also not possible to optimize the results by driving only when traffic is light. If taxis want to operate only when the average speed is greater (maybe from 10pm to 6am), they could lose out on revenue because the demand for trips at that time is relatively low.

If you see the information from a regulation or traffic perspective, it can help NYC regulate the amount of taxis that can be at service at a particular time of the day or see if there are bus routes that need to be created. In the long run this perspective can help the city reduce carbon monoxide emissions.

Cost Model

The cost model is available online (https://mavaladezt.shinyapps.io/taxis/) and can be modified in order to make simulations and understand how different variables might affect the profitability of the taxi business.

About the Data

Data of Jan-Jun 2019 trips was obtained from NYC Taxi and Limousine Commission (https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page). Only transactions paid with credit card or cash were used and it was assumed that 1 passenger was traveling when the driver was not able to introduce the number of passengers when the trip started.

About Author

Mario Valadez Trevino

Mario Valadez Trevino is a NYC Data Science Fellow with a B.S. in Industrial Engineering with minor in Systems Engineering and an MBA. Mario has relevant experience in demand forecasting, production and transportation planning, warehouse management systems and...
View all posts by Mario Valadez Trevino >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI