Metrolinx: Presto Fare Card

[nfitz]

Tap with your debit card, have TTC charge your VISA ($20 authorization because you ride frequently; an actual single charge will be made at the end of the week for all travel), and later confirm you paid for a fare enforcement officer by tapping your American Express. The card, for the purpose of the transit system, can be identification of the individual person and their back-office (Presto) account. It does not need to also represent the method of payment. There may be many registered ways of identifying a person, including a unique NFC phone code in a TTC app.

Doesn't that create potential abuse with daily fare capping where person 1 travels to work at 8 AM every day, and their partner then uses another of person 1's cards at 9 AM to take the kids to school, and head to work. Even if a transfer system is in place without daily fare capping, it would be hard to detect what's the start of a new trip, and whats a transfer, if both were travelling simultaneously in different vehicles.

I suppose it would work if you eliminated transfers and had not daily weekly, or monthly maximums.

 

[mpd618]

And if the system is moderately intelligent they can all be registered to identify the exact same person and any card will work just as well as any other card (including paying through the persons preferred method of payment).

If you're going to require registration of the card, then there's not that much value to the open-payment aspect of an open-payment system.

 

[Evening]

This sounds very interesting. Is this what David Miller wanted to implement instead of PRESTO because even as a PRESTO supporter, I gotta say this is much better. The only problem I see is the no fines from fare enforcement. Wouldn't that just encourage people to not pay and only pay when they get caught? Maybe instead of charging the $3 dollars at the end of the day, it charges a $100 fine. But I'm not sure how that works in legal terms (right to trial and all that).

I guess there is also an issue for people who don't have debit cards but is that really an issue today? And even if it was, we can have a presto like card as back up in this kind of a RARE situation.

Maybe we should have waited for this system instead of jumping onto PRESTO though anything is better than the TTC's status quo.

 

[rbt]

Doesn't that create potential abuse with daily fare capping where person 1 travels to work at 8 AM every day, and their partner then uses another of person 1's cards at 9 AM to take the kids to school, and head to work. Even if a transfer system is in place without daily fare capping, it would be hard to detect what's the start of a new trip, and whats a transfer, if both were travelling simultaneously in different vehicles.

I suppose it would work if you eliminated transfers and had not daily weekly, or monthly maximums.

The result is similar to today with metropass sharing with a small process in the background which looks for fraud (this is necessary regardless to detect and block passbacks). Having multiple cards in use in different locations where physical constraints would make it impossible would be a sign of possible fraud and should result in a second charge; such as using gate X at Finch then 30 seconds later (without tapping out) using gate Y at Finch.

 

[rbt]

If you're going to require registration of the card, then there's not that much value to the open-payment aspect of an open-payment system.

Who said require? Some infrequent user making a single trip obviously will not be registering. These users aren't going to get confused as to which card they used to pay; they'll tend to clearly remember as it's not something they do very often.

The semi-frequent riders attempting to hit the discounted-fare level for the month but went on vacation from the 5th to the 15th (can't remember what they used from 1st to 4th) are the ones who will forget what card they used. They're also going to be registering for the federal income tax deduction.


Of course the Presto backend, thus far, is not moderately intelligent. Heck, the backend doesn't even allow a single pooled fund for households today. That kind of thing should only take a couple days for a single person to implement if considered early in the design; Accenture isn't known for giving design advice which results in flexibility and reduced change-fees. Singe Presto seems to have an iterative design process, it really should have been an in-house team and not an outside contractor.

 

[rbt]

This sounds very interesting. Is this what David Miller wanted to implement ...

AFAIK the TTC was looking at something closer to what Vancouver/Salt Lake City have implemented. A credit/debit card would be a full single fare and a special purpose "Bank of TTC" card would be available for bulk-rate discounts. The technical effort on these systems to give bulk-rate discounts for a single open payment card is very low (including pre-authorizing additional travel to reduce transaction fees).


London's recent Oyster changes are heading in the direction I've described. By pushing most of the bulk-trip discounting into the backend and making the hardware the customer has (expected 2016) in their pocket much less intelligent, for identification purpose only. I suspect, if phone makers can improve NFC transaction speeds, they'll go to a phone app shortly after 2016 and begin eliminating the physical Oyster card entirely just to get rid of the card deposit fee complaints.

... instead of PRESTO because even as a PRESTO supporter, I gotta say this is much better. The only problem I see is the no fines from fare enforcement. Wouldn't that just encourage people to not pay and only pay when they get caught? Maybe instead of charging the $3 dollars at the end of the day, it charges a $100 fine. But I'm not sure how that works in legal terms (right to trial and all that).

Whatever fines and system put into place to prevent pass-backs on the subway portion (which is not POP) is what you would apply to other forms of fare-fraud. Pass-backs are one of the most common types of fare-fraud and one of the easier to detect once a tap-in is required at all gates (open-gates with TTC staff standing at them will have to disappear).

In the past on GO, you would receive a warning for forgetting to punch your ticket (they'd ask that you punch it on the way out). However, if the fare-enforcement officer recognized you or you had writing on your ticket indicating a second office, then they would enforce the fine.

I would propose configuring a first offence for TTC fare-fraud to charge a normal second fare and send the customer an email stating detection of suspicious activity, giving them a chance to explain/complain/report a card stolen, and notify them that the next time there would be a $100 fine. Usually for this kind of thing (not with transit but online store backends) I configure to allow a few free events per year (normal price) to let accidents slide. Habitual/frequent problems are addressed more sternly.

 

[muller877]

If you're going to require registration of the card, then there's not that much value to the open-payment aspect of an open-payment system.

Registration helps the TTC better understand the customer AND reduces cost. An open payment system may slightly increase the cost and/or time for approval of a transaction.

The TTC can become very precise about load counts and where there should be additional (or less) service with registration. The cost reduction comes from bulk purchases of services via the web (versus a kiosk or employee interface at the station).

I agree that not all cards should require registration...but there should be a penalty (or, alternatively an incentive) for registration. Without registration it should be the cash fare as it is now. With, the token fare (and the potential to have weekly or monthly maximums).

 

[mpd618]

Who said require? Some infrequent user making a single trip obviously will not be registering. These users aren't going to get confused as to which card they used to pay; they'll tend to clearly remember as it's not something they do very often.

I don't agree that someone who won't want to register is necessarily an infrequent user. If the transit agency says "you can use any contactless card to pay", many people will avoid getting a special purpose card.

The problem of what happens when you tap in with one card and tap out with another card is not faced by many transit agencies, I don't think. Most scenarios are either tap once, or separate entry and exit past turnstiles that won't accept a mismatch between cards used to tap in and tap out.

 

[TOareaFan]

Wouldn't the tax credit be incentive enough for frequent/regular users to register whatever card they are using for payment?

 

[muller877]

The result is similar to today with metropass sharing with a small process in the background which looks for fraud (this is necessary regardless to detect and block passbacks). Having multiple cards in use in different locations where physical constraints would make it impossible would be a sign of possible fraud and should result in a second charge; such as using gate X at Finch then 30 seconds later (without tapping out) using gate Y at Finch.

Another company has implemented a contactless system that we do not normally think about as a transit operator but it works in a similar fashion:

http://www.whistlerblackcomb.com/tickets-and-passes/rfid-access.aspx

It cost them than $10M...for the ERP software and the RFID readers ($12M in total capex for the year which includes this and new gondola's for the main lift). Of course this is for about 100 readers (37 lifts) for chairlifts and probably 200 more throughout the village and mountain for payments.

To prevent fraud everyone has a picture taken. The operators have a tablet which shows your face. (I'm sure a face recognition program could be used to reduce the need for human confirmation)

It operates multiple ways. You either can buy a season pass, a prepaid amount of days (which counts down when you use it) or have your credit card on file and it automatically charges it when you use the RFID chip. At stores you can tap it just like a credit card. (some resorts also add in the option for hourly purchases)

The readers on the mountain are large and convenient. Just put your card in your right hand pocket (and you wallet in the left) and it will read it without needing to remove it. And very fast approvals. The readers at stores are more like the credit card tap machines.

It records every RFID terminal you tap on and you can access the information on your smartphone right away. This helps search and rescue when they can't find a person (and in case there is a dispute).

It has all the bells and whistles that any transit system needs. I wonder which company implemented it for them? The only question in scalability. Could it be used with 10,000 devices vs 300?

 

[rbt]

It has all the bells and whistles that any transit system needs. I wonder which company implemented it for them? The only question in scalability. Could it be used with 10,000 devices vs 300?

10+ years ago hitting TTCs transaction and response time requirements with an off-the-shelf DB was tricky and expensive. Now it's almost trivial to hit 250k transactions per second on modest computer hardware using open source software (like PostgreSQL). Provided the back-office doesn't do anything foolish, scaling up the backend to fit the TTC requirements isn't going to be technically difficult.

The more challenging component is getting the network reliable even when some random utility decides to cut up a street, but that applies to nearly any implementation and I hope they solved that when they did the security camera installations.

 

[Catenary]

Another company has implemented a contactless system that we do not normally think about as a transit operator but it works in a similar fashion:

http://www.whistlerblackcomb.com/tickets-and-passes/rfid-access.aspx

It cost them than $10M...for the ERP software and the RFID readers ($12M in total capex for the year which includes this and new gondola's for the main lift). Of course this is for about 100 readers (37 lifts) for chairlifts and probably 200 more throughout the village and mountain for payments.

To prevent fraud everyone has a picture taken. The operators have a tablet which shows your face. (I'm sure a face recognition program could be used to reduce the need for human confirmation)

It operates multiple ways. You either can buy a season pass, a prepaid amount of days (which counts down when you use it) or have your credit card on file and it automatically charges it when you use the RFID chip. At stores you can tap it just like a credit card. (some resorts also add in the option for hourly purchases)

The readers on the mountain are large and convenient. Just put your card in your right hand pocket (and you wallet in the left) and it will read it without needing to remove it. And very fast approvals. The readers at stores are more like the credit card tap machines.

It records every RFID terminal you tap on and you can access the information on your smartphone right away. This helps search and rescue when they can't find a person (and in case there is a dispute).

It has all the bells and whistles that any transit system needs. I wonder which company implemented it for them? The only question in scalability. Could it be used with 10,000 devices vs 300?

I'd put money on the system being run by RTP|ONE with equipment from SKIDATA.

http://www.activenetwork.com/solutions/rtp-one


http://www.skidata.com/en.html

SKIDATA makes some public transport related equipment.

 

[NotMathematical]

New PRESTO readers for the UP Express.
From the torontonist: http://torontoist.com/2015/03/riding-the-union-pearson-express/slide/20150309-torontoist-up-201/

 

[Catenary]

New PRESTO readers for the UP Express.
From the torontonist: http://torontoist.com/2015/03/riding-the-union-pearson-express/slide/20150309-torontoist-up-201/

That's the Scheidt & Bachmann SV|50 which I mentioned in this post.

It appears to have the cutout for the barcode reader, but it's hard to tell if there is one installed. It's also interesting that the buttons aren't labelled for anything.

The bigger ticket vending machines also appear to have a contactless card reader below the screen. It's not PRESTO labelled, but it's not the type for debit or credit cards.

 

[cplchanb]

Im curious to know if there will be any type of quota enforcement for the trains. Whos to stop the trains to be loaded beyond seated capacity?