The dataset is provided by the MTA (Metropolitan Transportation Authority) and is owned by NY Open Data. As part of the Subway Action Plan, the MTA started to release a series of Customer Journey-Focused Metrics to better measure the rider experience, made possible by investments in new train tracking technology and more robust methods for determining how riders use the subway system.
This project looks at the Customer Journey Time Performance (notated as ‘JTP’ in this project) which is the estimated percentage of customers whose journeys are completed within 5 minutes of the scheduled time, reported each month and on each line.
I have reviewed the data and focused my analysis on the data from 2023 and the Q & B train lines. These train lines are my main source of transportation to and from my home in Brooklyn, NY. During my analysis, I focused on the following questions:
1. Across all lines, does an increase to the number of passengers also increase the ‘additional platform time’ for riders?
2. Which division of lines have the greatest number of passengers?
3. For the Q line, as the additional platform time (‘apt’) rises, how does that affect the Customer Journey Time Performance?
4. What is the worst and best month for Customer Journey Time Performance of the Q line?
5. How does the Q train perform during off-peak and peak times of day?
6. What are the top 5 performing lines of 2023? And what are the worst performing lines of 2023?
7. By month, what line has the most affected customers due to poor performance?
8. What is the average Additional Platform Time from worse to best?
Answers to the above questions (not intended for a non-technical audience or for stakeholder presentation purposes, analyst use only):
Across all lines, does an increase to the number of passengers also increase the ‘additional platform time’ for riders?
2. Which division of lines have the greatest number of passengers?
3. For the Q line, as the additional platform time (‘apt’) rises, how does that affect the Customer Journey Time Performance?
We can see from these visuals that as the additional platform time ('apt') increases the Journey Time Performance ('jtp') decreases, as expected.
4. What is the worst and best month for Customer Journey Time Performance of the Q line?
The above visual shows us that the Q train line has the best performance in January 2023 (Peak times) and the worst performance in September 2023 (Peak times). This could be due to lower tourism in January which leads to a decrease in ridership and track maintenance causing delays in September. A further analysis of why the performance was high or low would require further research.
5. How does the Q train perform during off-peak and peak times of day?
Using the previous visual, there are many months where the Q train performs better on JTP during peak times than during off-peak times. This may be due to track maintenance projects during the daytime off-peak hours causing delays, but the "why" would need further research.
6. What are the top 5 performing lines of 2023? And what are the worst performing lines of 2023?
7. By month, what line has the most affected customers due to poor performance?
8. What is the average Additional Platform Time from worse to best?
So we see here that the B train has the longest average additional wait time, almost 2 minutes extra from the scheduled departure time. So the B train may not be that express if you are looking to avoid delays. The best lines for APT are the shuttle (S) trains and the L train, all under one minute wait.
Conclusion: We are now able to draw some conclusions and insights into our initial questions: Is the B train really express and is the Q train reliable for this NYC local?
The B train has the longest additional platform time average out of all the lines of NYC. So, if you're looking to hit that express track, you may have a slower than expected journey.
The Q train performance may not be as reliable as this NYC local would prefer. But during peak times, the train seems to perform well. But during the off-peak times, there is a large probability that I will experience a schedule delay.
