The Data School - When and Where Should You Be a Taxi Driver

Today we were provided with data about Yellow Taxi Rides from the NYC Taxi and Limousine Commision. As a child my dream job was to always be a taxi driver, but my mother wanted me to be a doctor so we compromised and I became a data consultant. As I started examining the data a question came to mind:

When would be the best time to be a seasonal taxi driver?

The first challenge was to convert the data from the PARQUET format it came in to CSV. Using python and a few lines of code I was able to get all 12 months of data into a format that Alteryx could understand.

The next challenge was dealing with the size of the data. After unioning all 12 months of data I was left with a file that was nearly 4GB in size. My favourite tool in alteryx is the autofield tool which automatically changes each fields datatype to the most efficient one possible. By running this tool I was able to diminish the file size from 4GB to 1GB.

There was also a shape file available providing a map structure for the different pickup and drop off locations in NYC. I wanted to show different routes in Tableau, so in order to create geospatial drop off and pick up points I have to find the centroid of the polygons.

Now that I have my data I have to figure out how I am going to visualise. I wanted this dashboard to focus on where and when.

Using coolors I was able to use this neat open source image of a taxi to make a color palette. And using a color wheel provided by canva I picked complementary colors to the main yellow color.

<color-palette name="Taxi" type="regular">

<color>#FFFFFF</color>

</palette>

The final dashboard I came up with didn’t include routes because I was having trouble getting them to map properly. But I did find out that the most profitable time to be a taxi driver is in September driving at 5am to and from Newark airport. Check it out here.