Closed novalis closed 7 months ago
Related GTFS proposal: http://groups.google.com/group/gtfs-changes/browse_thread/thread/1416890017239faf
--novalis
That GTFS link is super helpful. I am just going to paste the relevant text into the ticket.
Proposal:
co2_per_km (optional): The co2_per_km field allows the agency to specify how many grams of carbon dioxide are emitted for each kilometer that a passenger travels on this route. The value should be an integer, and should reflect an individual passenger's "share" of the emissions, rather than the total emissions of the vehicle. If no emissions information is available, the field should be omitted or left blank (as a value of "0" indicates
Discussion: This is a value that's controversial and tricky to calculate. Ideally, a passenger's "share" of the emissions would reflect how many passengers are on the vehicle at that point, but realistically, most applications aren't going to have such detailed information. Providing a per-route average allows the agency to incorporate the factors that they think are important based on their knowledge of their operations.
Your thoughts?
Joe Hughes Google
--mkeating
the environmental benefits of using their systems and make that information available to the public and agency management.
An interesting idea. Just resurfacing this 10-year old thread with a related thought or two.
First, I noticed that the 1-2 responses to Joe's idea from the mailing list suggest that highlighting CO2 emissions is (a) problematic technically and (b) might not show diesel transit buses as being a very environmentally appealing option:
my memory of diesel bus energy efficiency is approximately 33 passenger-miles-per-gallon. This is worse than 2 people in a car with average fuel efficiency. A nearly full bus with 50 passengers would typically get 200 passenger-miles-per-gallon, much better than anything else but a full Prius
That said, now that some time has passed, some other thoughts/ideas:
1) Agencies are now (finally?!) starting to pick up (battery-) electric buses. Perhaps highlighting the emissions of the bus is now a good idea, as it could help people get excited about (much lower emissions) transit vehicles.
2) Previously, the lack of car routing was a blocker. But that's done now right?
3) If the purpose is to inform users of the potential CO2 savings from taking the bus, perhaps the complexity (of calculating bus emissions or marginal passenger emissions) noted on the mailing list should be completely sidestepped by instead just calculating the average passenger vehicle emissions, and presenting it like: "by taking the bus, you could be keeping around __ [quantity] of CO2 in the ground". As some folks noted, since the bus is driving the route anyway, you can assume its emissions are fixed (whether or not the user becomes a passenger) and just focus on the SOV passenger emissions prevented.
4) To "make that information available to the public and agency", perhaps the planner could keep a record of all trips planned, and provide some aggregate statistics. E.g. "The number of transit trips planned this month represent a potential savings of ___ [quantity] of CO2 from all passengers".
At Entur we are probably doing something on this in the coming year, but it might happen as a decorating service on top of OTP.
The key question for OTP is: Should it be possible to search for travel plans with CO2 emissions as part of the criteria. We will probably not do that, and just decorate the itinerary legs with co2 emission estimates. Average estimates pr passenger seems like the best option here. But, this needs discussion - OTP should just provide the data, and then the clients can handle the presentation.
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
+1 for handling the data in OTP and letting clients take responsibility for presentation.
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
One of digitransit users suggested integrating a data source, which describes emission estimates per agency and per transport mode, into OTP. The data would include default values in case some agency does not provide the requested data.
The idea is that some agencies operate their bus lines using modern electric vehicles whereas others may use older technology. Agency specific data will make estimates more accurate.
@miles-grant-ibigroup and @daniel-heppner-ibigroup have implemented this in the IBI frontend. They may be able to contribute their expertise.
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
We are planning to implement something soon, lets discuss this in today's dev meeting.
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days
We have implemented a sandbox in https://github.com/opentripplanner/OpenTripPlanner/pull/5278 https://github.com/opentripplanner/OpenTripPlanner/pull/5489 https://github.com/opentripplanner/OpenTripPlanner/pull/5505 . I'll close this issue now and if someone wants to improve the functionality and/or add support for a some input data format, we can create a new issue about it.
I have heard a few transit people talk about the need to calculate the environmental benefits of using their systems and make that information available to the public and agency management. A step toward that would be for the trip planner to tell you how much CO2 you will have avoided emitting by taking transit for a particular trip instead of driving a typical car. This would be pretty easy to calculate, and perhaps other trip planners already do it. Just a thought for later on.
Nick BS' comment: Interesting idea. This is definitely doable, assuming we can start with some pretty broad assumptions (e.g., CO2 emitted/mile driven is constant). Also, to get good results we'd need to have implemented car routing, which isn't really on our roadmap. Or maybe we could get away with pretending that a car would have been able to just drive as a bike would?