|
|
| Re: More Than You Want to Know About MTA's Ridership Modeling (1631625) | |||
 |
|||
| Home > SubChat | |||
|
[ Read Responses | Post a New Response | Return to the Index ] |
|
||
|
Re: More Than You Want to Know About MTA's Ridership Modeling |
|
|
Posted by Stephen Bauman on Tue Apr 23 23:21:21 2024, in response to More Than You Want to Know About MTA's Ridership Modeling, posted by heypaul on Tue Apr 23 20:22:18 2024. I read that the MTA will be making this making this data available to the public. Its usefulness will depend on how they accumulate the data.They used to publish turnstile data. Unfortunately, they followed what they had done before computers. Totals for each turnstile bank were read every 4 hours. Even worse, the 4 hours varied from station to station - even along the same line. This made it fairly useless for determining where and when crush loads are likely to appear on a platform. This data also included the exit counter. The people now in charge of complying with the MTA's open data requirements stopped posting this data. The replacement gives hourly entrance totals by station - not turnstile bank. Nobody cares when/where people get off. There are devices that can accurately measure how many passengers are entering and leaving trains or buses. These are cameras that are mounted at the doors. The software can determine the whether passengers are getting on or off. It can process the numbers in real time. System-wide use of these devices could provide the basis for an accurate passenger use metrics. There are weight sensors on each rail car. The weight sensors could be used to estimate the passenger count in each car. This is what the passenger loading displays use on the LIRR. It was implemented internally by the MTA. Operations and planning were not interested in this data. It was added as passenger information enhancement, like the count down clocks. There's no reason why this could not be extended to the subways. Some existing real time GTFS data for MTA buses has a field for the number of passengers on the bus. There are problems with using the bus GTFS-RT data because of how it's collected. The device on the bus will transmit every 30 seconds with information it knows: the vehicle number; vehicle position (lat/long); and passenger count. It will transmit this information, only if it has moved within the last 30 seconds. The MTA must then figure out the trip id, route and direction from the raw data. That's a bit complicated. However, I think using measurements on the vehicles themselves, rather than passenger data (turnstile or OMNI taps) should prove a better solution. The problem is that the MTA's existing compression methods for the vehicle data is embedded in its culture. |
|
|
|