Subway riders cringe when a delay is announced. Emotions range from mild annoyance to outbursts of anger. Delays inconvenience us and reduce productivity. By default we blame subway management (TTC in Toronto). Is this fair? Wouldn’t it be nice if we had data to determine what causes the delays?
Museum station in Toronto.
Toronto Star Article
This headline caught my attention: “What’s keeping that train? The TTC has 69,000 reasons why”.
I’ve been using the subway since 2000. I’ve been curious to learn more about TTC (subway) delays.
The transit agency receives so many requests for data it finally decided to make the statistics available on the 69,000 delays from January 2014 to April 2017.
Shocking highlights:
“Palmer said people go onto the track for different reasons, including to urinate, to retrieve something they dropped, or on a dare.”
What? I can understand accidentally falling onto the tracks but intentionally doing it? It gets worse..
That could be urine, it could be feces, it could be vomit,” said Palmer. “New Year’s Eve…we had extra crews on just cleaning up vomit and nothing else. Pretty much every train that went to Finch had vomit in one or more cars.”
The majority of us are civilized. It’s a tiny minority of riders that cause these completely unnecessary delays. Other ‘people’ delays are caused by unforeseen health issues and injuries. Aren’t you glad that the TTC stops the subway to look after us? There are also delays due to maintenance issues.
The Star article summarized the issues nicely at the individual delay level but I was curious about the higher level categories. Let’s take a look at the data.
The Data
Original Data
- Subway&SRT Logs(Jan01_14 to April30_17).xlsx main data set. 69043 rows, 3.5 MB
- Subway&SRT Log Codes.xlsx delay codes and description
Note: ‘Subway‘ is the main network. ‘SRT‘ is a smaller line (Scarborough Regional Transit).
My Excel File
I combined the two main Excel files added formulas and pivots: my Excel file. (10MB Excel file)
Delay Categories
I wanted to see the results at a higher level so I created these categories:
Passenger, Mechanical, Debris, Maintenance, TTC employee, TTC management, Weather, Fire Smoke
Sheets in my mini model:
- Kevin’s Delay Categories Detailed delay codes will map up to these 9 high level categories
- Subway&SRT Delay Codes I selected high level category for each delay code (columns E & J)
- Delay Data Original data in columns A to J. Formulas to determine mapping (columns K to M)
- Pivot Tables I role delay codes up to my 9 categories (count & time) to get bird’s eye view!
The Results
After mapping individual delay items to my categories these two pivots tell us a lot:
The ‘Passenger’ category is the highest for both ‘Count of Delays’ and ‘Time of Delays’. It’s still important to examine and improve all other delay categories, but we shouldn’t blindly blame the TTC for everything. If you run to catch the train and get stuck in the door that is a delay affecting thousands of people and you are part of the data!
The data reveals a lot of details about what really goes on inside the subway. There are a lot of people working very hard to (a) keep us safe and (b) improve the TTC. I’ve been personally guilty of blaming the TTC when delays are often caused by us the passengers. The TTC isn’t perfect and we need more funding for maintenance improvements and expansion, but over the past few years they have focused on customer service and reduced delays.
Subway Expansion
Speaking of Subway expansion…the Yonge line extension to Vaughn opened yesterday, December 17th 2017. The TTC offered free rides all day but I decided to stay home with my wife and our dogs. The only traffic we caused was to Netflix’s servers 🙂
Andy Byford
As mentioned above, the TTC has been focusing on customer service and reduced delays. After finishing my mini-analysis in Excel I started reading articles related to TTC improvement efforts.
I wondered who was behind these efforts. Various articles point towards departing TTC CEO Andy Byford. Mr Byford must have done something right as he is now headed to New York city to be President and CEO of New York City Transit authority. Looks like he’ll be very busy! Further reading:
- i-stood-up-for-the-ttc-andy-byford-reflects-on-his-tenure
- andy-byford-stepping-down-as-head-of-ttc
- ttc-named-best-public-transit-agency-in-north-america
- Failing Subway Threatens New York’s Financial Future
- nyregion/subway-delays-lost-work-time-cost-new-york
- outgoing-TTC-CEO-Andy-Byford-goodbye-message
2017 APTA Transit System of the Year
Byford’s 5 year modernization plan appears to be the reason for winning the award. Some have scoffed at the TTC receiving this award this year but most people have only used 1 subway system so how could they compare the TTC to other subway systems? Let’s investigate this a bit further. Here is the link for APTA. I didn’t take me long to find their data!
About Me
I have taken the TTC (Toronto subway) thousands of times over the past 18 years. When headed out of the office I take the subway from Museum station to Union station. I know exactly which subway car and door will get me closest to the stairs at Union where I dash off to catch my GO train home.
Fascinating! Great blogpost.
Thanks Oz. It was interesting to study the data with Excel. The TTC has done a great job reducing the delays.