To assist the operation team of the NYC Taxi commission in Manhattan to optimize the distribution of the taxi fleet
- Download the code (Contact me if you want the token to download via linkedin)
- Install the requirement with this syntax : pip install -r requirement.txt
- Run the ipynb file in jupyter notebook
Business Problems : Operation team of the NYC taxi commission want to optimize the distribution of the taxi fleet in 1 month ahead to maximize the revenue.
To solve that problem, I make a time series forecasting model using prophet library. now let's see the dataset
Dataset have 19 columns with 4 columns have null value, but the columns that have null data is not important in our analysis, so we dont need to do anything from that columns.
Now let's make a timeseries model
- Prepare the data so that it is ready to be used to create a mode
- Take a important columns that you need, in this case we need pickup datetime and location and order quantity. (in my code the data was in variable called final)
- Choose one location that you want to analyze (in my case I Choose location with ID 4)
- Make a time series model
- Make a prediction
- Last, Analyze data from prediction
From the data above, we can recommend to Operation team of the NYC taxi commission to send about 15-20 taxis between 23:00 to 03:00 everyday and besides these hours it is enough to prepare 10-12 taxis