for years that during peak hours the should only have 2-3 buses stop from Forest Hills to The Square. Instead the 14, 30, 34, 34e, 35, 36, 37, 40, 51 and 50 all make every stop to and from The Square and The Hills. PS Add School Buses at 8am, absolutely absurd.
And don't get me started on the fact the Malden, Chelsea, Medford and Porter's Commuter Rail stop are considered zone 1A but Roslindale, West Roxbury and HP are all zone 1 but are with in city limits.
I agree that Washington gets bogged down with all those bus routes. However, it is extremely frustrating when you are taking a bus TOWARDS Forest Hills and empty buses pass you. For some reason, when it is busy in the morning, it seems like only the crowded buses will stop to pick you up.
Always thought that no bus that passes The Square should stop prior to reaching the Square, but instead have a bus that loops back and forth. However I think adding an additional bus in the effort to reduce traffic would be something that would require some in depth analysis. But seeing how populated that stretch is, I think having a designated bus would be a great service for the area, particularly for the elderly and disabled who are hard pressed to find seats. Also their are a lot of young mothers in that are who rely heavily on public transit, they are often unable to board packed busses with strollers.
But then again I'm no urban planner, just a life long resident putting in my two cents.
So to get from Downtown to anywhere past Roslindale Square you take the orange line to a bus, then transfer to another bus at a bus "hub" in the square. Tacking another 5-15 minutes onto everyones commute?
Maybe a better solution is to run the longer lines (ie 34e) as limited/express routes during rush hour, which the T already does to some extent.
would bypass Washington St until you passed the Square vs the current issue of EVERY bus stopping at EVERY stop until you reach the Sq.. It what creates most of the congestion.
Washington St from the square on down to FH is a clusterphuk. WAY too many stops.
Also, Roslindale, Bellvue, Highland,W. Roxbury, H.P. and Fairmount should all be Zone 1A. These city locations are not served by rapid transit lines and are deserving of the discount to 1A. Commuter rail stops inside the city limits should all be 1A, like rapid transit.
The above mentioned Boston stops are all further out from the central stations than West Medford is (only 4.5 miles from North Station at most), and we don't get rapid transit service in our direction just yet. Although I do agree that it is silly for Fairmount to not be included at the very least, given the lack of transit in that direction.
However, overall distance from South or North stations is why.
Not that South and North Stations should be considered the center of anything. One big general issue with the T is the assumption that everybody works downtown.
And I'm aware 'zones' are determined by miles, and that Roslindale, for example, falls out of the Zone1A while Medford and Porter Square [for example] are much closer to downtown Boston. My point is Boston should be able to strong arm the MBTA to make all commuter rail stops inside city limits 1A like rapid transit stops.
And yes, Fairmount should be 1A. Get on/off at the next stop, Morton St, and monthly pass drops from $176./mth to $70./month. A ridiculous differential. Hyde Park, Roslindale and the W. Roxbury stops should likewise be 1A. Bus congestion on Centre and Washington streets, and at Forest Hill station would DRAMATICALLY decrease.
It would make more sense to change the stations within the city of Boston to a City Zone, charging fares from $2.00 to $3.50 for stops within Roxbury, Roslindale, Hyde Park, Readville and West Roxbury. Convert Zone 1A to City Zone 1 and charge $2.00 single fare/$70 pass, while stations up to the city line (City Zone 2) charge $3.50 single fare/$110 pass, the same as an Inner Express bus pass.
IMHO, the reason the T charges $5.50 within the city is that the T makes more money, with less resistance, from commuter rail riders than they do bus and subway riders. That's the way the T likes it: bedroom community riders coming in, handing over their $173 per month pass, and getting to Boston faster than the riders making three bus connections, navigating choked bus and train aisles, sitting in traffic and at stations because the signals were acting up, or that some kid ran up and down the tracks, etc.
As you point out, it's faster and more comfortable. It's reasonable to charge more, but the fare escalation should be adjusted. I can board a Needham Line train at Forest Hills, and be at Back Bay 8 minutes later for $2.00. But if I board a mile to the south in Rozzie Square, I pay $3.50 for that extra mile, making a total fare of $5.50. That's about a dollar per mile. I'm not sure, but I doubt Providence passengers are paying remotely close to a dollar per mile.
Maybe Rozzie's CR fare should be lower. But the fare isn't *supposed* to be a flat rate per mile -- no transit system charges that way.
Anyway, I'm more concerned with the terrible schedule than the fare. It doesn't matter how cheap a train is if you have to wait 100 minutes for it, and there's no weekend service.
I've said it before: single-car single-employee DMUs every 15 minutes!
The idea sounds appealing, but there are no manufacturers that meet FRA standards. I suspect four to five shorter trains over the same time period would also prove difficult for South Station track management. The real fix for Rozzie and West Roxbury is an Orange Line extension. Messing around with DMU or EMU type solutions just pushes the problem back a hit, but fundamentally, the population and transit utilization of the area is of sufficient density to suggest rapid transit as the best option.
DMUs are financially feasible. Subway extensions aren't.
All the T has to do is buy DMUs the next time part of the commuter rail fleet is being replaced (cough, Hyundai-Rotem).
FRA waivers are now possible. The DCTA in Texas surprised everyone when their waiver was approved in 2012. It appears that the leadership at the FRA is finally acknowledging that preventing crashes with signals, and using energy-absorbing European designs, can be safer than requiring rigid overweight trains that nobody outside the U.S. would ever consider.
Subway extensions are so expensive that they're never going to happen. DMUs and the related service increases could happen virtually overnight on all the existing commuter rail lines.
And if Southwest can turn a plane in 15 minutes, and high-speed trains in Japan can turn in 5 (http://www.youtube.com/watch?v=wHznLv4ucCI ), I'm sure the T and Amtrak could figure out how to make better use of the South Station platforms if they tried.
In my long experience riding buses past Roslindale Square, they almost always lose half their passengers between FH and the Square. So if they took separate buses, the longer routes' passenger counts would plummet. The every-10-minute service along Belgrade Ave would probably drop to every 20 minutes or worse, for example. And would they really be much faster? They'd still have to deal with all the trucks, school buses, and cars on Washington St., not to mention all the buses that would still be picking up the crowds along the way.
isn't on the list because 9 other routes share the most crowded section of the route. Total ridership on all buses running between Rozzie Square and Forest Hills is more than enough to justify extending the Orange Line to Roslindale. We need to get the new Mayor involved in some transit advocacy for Southwestern parts of the city.
The elimination of certain stops seems like a good move, so I'll give the thumbs up on that.
The T has talked about bunching before, but nothing has happened. Unless they explain what they're going to do, I'm assuming nothing will happen this time either.
The solution seems pretty simple; just send the first bus of the bunch "express" from a terminus until it gets back on schedule, or at least properly spaced. Passengers who want to get off an an intermediate stop should simply board bus #2 of the bunch.
A fuller bus takes longer for people to get on and off, as they negotiate narrow aisles blocked by other riders.
When a bus starts to fall behind, the number of waiting riders at each stop grows further, which means that when a bus does finally stop, it takes even longer to board everyone.
Skipping/passing-up stops can space buses out a bit better, and the T does do this, but it doesn't solve the real problem.
It may look like the buses are doing better because they arrive "on time" at the next terminus, but you've passed up a whole bunch of people to do it. That defeats the point of bus service, which is to serve people. And the crowds at the passed-up bus stops will grow to the point that when a follow-up bus does stop, it will be overwhelmed.
To make the best of a bad situation, when a trip gets dropped due to delays, it may be better to just send the buses out together as a convoy to handle the additional waiting riders.
Prevention is the best cure: speed up boarding and alighting with level boarding and curb extensions; faster fare payment (or use proof of payment); signal priority for more predictability at traffic lights; fewer, but better designed, stops.
Supposedly the key bus route improvement project does some of this. We shall see.
No, my suggestion still helps. There is no benefit at all to running the buses bunched. Although your solution is "better," it requires infrastructure that simply will not be built in my lifetime in Massachusetts.
Think about it:
If bus #2 is right behind the expressed bus #1, then the folks waiting at the intermediate stops will not have to wait much longer than they otherwise would by not expressing running bus #1. The fact that bus #1 passes them without stopping is probably frustrating, but ultimately of little consequence, if bus #2 is only a minute or two behind.
The folks further down the line (i.e., those past the "expressed" stop) benefit greatly, as the bus get to them sooner, and hopefully closer to schedule.
So why shouldn't they do this, at least in the short term?
If bus #2 is right behind the expressed bus #1, then the folks waiting at the intermediate stops will not have to wait much longer than they otherwise would by not expressing running bus #1. The fact that bus #1 passes them without stopping is probably frustrating, but ultimately of little consequence, if bus #2 is only a minute or two behind.
I may have skipped a step and left you confused.
I am presuming that the reason bus #1 and bus #2 are leaving at the same time is because bus #1 was significantly delayed on its most recent trip to this terminus. Therefore, the subsequent trip is also delayed.
That means there is a longer-than-normal gap in frequency serving the bus stops. And during that longer gap, more riders arrived at the bus stops and started waiting. If there is enough delay, then the crowds at the bus stops could start to overwhelm a single bus, causing it to become delayed itself, and propagate the bunching backwards.
I have observed this in real life too, so I'm not just speaking theoretically here. Once in a while, I have found myself waiting for 30 minutes during rush hour at a key bus route stop, while being passed up by buses both full and half-empty. The full ones do not stop because they were at crush load. And the half-empty ones do not stop because they were trying to catch up to their schedule (presumably). The more frequently this happens, the more likely that the route is overloaded.
For example, Vancouver actually keeps track of pass-ups. They also have some of the busiest bus lines on the continent.
I've long said to my friends that "the 73 stops every 200 yards." Finally the map of that route shows I'm not far off, at least for the part of the run in between School and Mt. Auburn Sts. I'm glad they're eliminating stops. There's no reason to have them that closely spaced. (Which might have worked in the days when it was a trolley route, maybe?)
No so fast, how do we know the former bus stops will turn into parking for cars? Maybe we'll get Hubway stands or private bike parking in there. Or maybe the former stops will continue to be officially off-limits for parking and fill up with the private cars of the well-connected. Maybe, as long as I'm dreaming, we'll get some bus-sized mini parks or food truck zones.
We don't need to dedicate more land to bike parking.
I've never had trouble finding a place to lock my bike. Except in Back Bay, but the problem there isn't a lack of land -- it's not enough posts to lock to, because of the multispace parking meters.
If you look at the plans for route 1, they are expanding the length of many stops by removing 1-2 adjacent parking spots. I think this will probably lead to a net loss of spots, as the number of stops being removed altogether is small.
Thanks ckd! The #1 bus route is slated to lose 15 stops gain 62 parking spaces net, a bit more than four parking spaces per bus stop on average. I'd still love to see some creative use of some of the reclaimed space.
But IMO the biggest causes of delays are fare collection, and badly programmed traffic lights. Removing some bus stops won't solve these problems.
For the first problem, they should make it easier to put a 7-day pass on a CharlieCard. And stop encouraging people to do the stupid value-loading procedure on the bus.
The whole point of the cash/ticket surcharge was to encourage people to use a faster means of payment. But what's the point, when they give the discount when you load money onto your card on the bus, which takes *longer* than just dropping in the cash?
I think they should eliminate the cash surcharge, and stop letting people add cash to cards on the bus. Using a CharlieCard should be its own reward.
And they should admit that the farebox coin and bill slot design is a failure. Next time, they should go with industry standard fareboxes, where you can dump in a handful of coins at once, and bills scroll by a window for the driver to accept rather than having a computer always reject them.
The traffic light problem would require some more effort and money. But that's easy -- the hard part is just getting the people in charge to make it a priority.
Comments
I'll believe it when I see
I'll believe it when I see it.
there's nothing quite like seeing a block's worth of bus
Some commutes, you can see 3 39 busses bumper-to-bumper. It's hilarious - they fill an entire block in JP. Looks like a giant train.
Why is it that the BOSTON globe never covers this stuff?
The #39 through Mission Hill and Northeastern is insane
It's as bad, if not worse than the B Line through BU. At least half the stops could be gone tomorrow and everybody would be better off.
The T should do that, add signal priority at Brigham Circle and run 60' buses 100% of the time. The improvement would be dramatic, to say the least.
The best part is...
...the T project page, which includes updates on the status of each route.
The "design status" is "complete" for all of the changes except route 66, which is merely "conplete." They'll get that extra hump soon, I'm sure.
I'v been saying
for years that during peak hours the should only have 2-3 buses stop from Forest Hills to The Square. Instead the 14, 30, 34, 34e, 35, 36, 37, 40, 51 and 50 all make every stop to and from The Square and The Hills. PS Add School Buses at 8am, absolutely absurd.
And don't get me started on the fact the Malden, Chelsea, Medford and Porter's Commuter Rail stop are considered zone 1A but Roslindale, West Roxbury and HP are all zone 1 but are with in city limits.
I agree that Washington gets
I agree that Washington gets bogged down with all those bus routes. However, it is extremely frustrating when you are taking a bus TOWARDS Forest Hills and empty buses pass you. For some reason, when it is busy in the morning, it seems like only the crowded buses will stop to pick you up.
I'v also
Always thought that no bus that passes The Square should stop prior to reaching the Square, but instead have a bus that loops back and forth. However I think adding an additional bus in the effort to reduce traffic would be something that would require some in depth analysis. But seeing how populated that stretch is, I think having a designated bus would be a great service for the area, particularly for the elderly and disabled who are hard pressed to find seats. Also their are a lot of young mothers in that are who rely heavily on public transit, they are often unable to board packed busses with strollers.
But then again I'm no urban planner, just a life long resident putting in my two cents.
So to get from Downtown to
So to get from Downtown to anywhere past Roslindale Square you take the orange line to a bus, then transfer to another bus at a bus "hub" in the square. Tacking another 5-15 minutes onto everyones commute?
Maybe a better solution is to run the longer lines (ie 34e) as limited/express routes during rush hour, which the T already does to some extent.
No you
would bypass Washington St until you passed the Square vs the current issue of EVERY bus stopping at EVERY stop until you reach the Sq.. It what creates most of the congestion.
Agree 100%
Washington St from the square on down to FH is a clusterphuk. WAY too many stops.
Also, Roslindale, Bellvue, Highland,W. Roxbury, H.P. and Fairmount should all be Zone 1A. These city locations are not served by rapid transit lines and are deserving of the discount to 1A. Commuter rail stops inside the city limits should all be 1A, like rapid transit.
Pps
Sorry for the typos, using my crackberry
City Limits are Irrelevant for MBTA
The above mentioned Boston stops are all further out from the central stations than West Medford is (only 4.5 miles from North Station at most), and we don't get rapid transit service in our direction just yet. Although I do agree that it is silly for Fairmount to not be included at the very least, given the lack of transit in that direction.
However, overall distance from South or North stations is why.
Not that South and North Stations should be considered the center of anything. One big general issue with the T is the assumption that everybody works downtown.
I'm the OP
And I'm aware 'zones' are determined by miles, and that Roslindale, for example, falls out of the Zone1A while Medford and Porter Square [for example] are much closer to downtown Boston. My point is Boston should be able to strong arm the MBTA to make all commuter rail stops inside city limits 1A like rapid transit stops.
And yes, Fairmount should be 1A. Get on/off at the next stop, Morton St, and monthly pass drops from $176./mth to $70./month. A ridiculous differential. Hyde Park, Roslindale and the W. Roxbury stops should likewise be 1A. Bus congestion on Centre and Washington streets, and at Forest Hill station would DRAMATICALLY decrease.
Convert city commuter rail stations to City Zone fares
It would make more sense to change the stations within the city of Boston to a City Zone, charging fares from $2.00 to $3.50 for stops within Roxbury, Roslindale, Hyde Park, Readville and West Roxbury. Convert Zone 1A to City Zone 1 and charge $2.00 single fare/$70 pass, while stations up to the city line (City Zone 2) charge $3.50 single fare/$110 pass, the same as an Inner Express bus pass.
IMHO, the reason the T charges $5.50 within the city is that the T makes more money, with less resistance, from commuter rail riders than they do bus and subway riders. That's the way the T likes it: bedroom community riders coming in, handing over their $173 per month pass, and getting to Boston faster than the riders making three bus connections, navigating choked bus and train aisles, sitting in traffic and at stations because the signals were acting up, or that some kid ran up and down the tracks, etc.
Commuter rail should cost more than subway, but zone 1 is crazy!
As you point out, it's faster and more comfortable. It's reasonable to charge more, but the fare escalation should be adjusted. I can board a Needham Line train at Forest Hills, and be at Back Bay 8 minutes later for $2.00. But if I board a mile to the south in Rozzie Square, I pay $3.50 for that extra mile, making a total fare of $5.50. That's about a dollar per mile. I'm not sure, but I doubt Providence passengers are paying remotely close to a dollar per mile.
Maybe Rozzie's CR fare should
Maybe Rozzie's CR fare should be lower. But the fare isn't *supposed* to be a flat rate per mile -- no transit system charges that way.
Anyway, I'm more concerned with the terrible schedule than the fare. It doesn't matter how cheap a train is if you have to wait 100 minutes for it, and there's no weekend service.
I've said it before: single-car single-employee DMUs every 15 minutes!
DMU service has other issues
The idea sounds appealing, but there are no manufacturers that meet FRA standards. I suspect four to five shorter trains over the same time period would also prove difficult for South Station track management. The real fix for Rozzie and West Roxbury is an Orange Line extension. Messing around with DMU or EMU type solutions just pushes the problem back a hit, but fundamentally, the population and transit utilization of the area is of sufficient density to suggest rapid transit as the best option.
DMUs are financially
DMUs are financially feasible. Subway extensions aren't.
All the T has to do is buy DMUs the next time part of the commuter rail fleet is being replaced (cough, Hyundai-Rotem).
FRA waivers are now possible. The DCTA in Texas surprised everyone when their waiver was approved in 2012. It appears that the leadership at the FRA is finally acknowledging that preventing crashes with signals, and using energy-absorbing European designs, can be safer than requiring rigid overweight trains that nobody outside the U.S. would ever consider.
Subway extensions are so expensive that they're never going to happen. DMUs and the related service increases could happen virtually overnight on all the existing commuter rail lines.
And if Southwest can turn a plane in 15 minutes, and high-speed trains in Japan can turn in 5 (http://www.youtube.com/watch?v=wHznLv4ucCI ), I'm sure the T and Amtrak could figure out how to make better use of the South Station platforms if they tried.
You're
Irrelevant, you don't live in these effected areas.
And Readville!
And Readville!
Sounds good, but...
In my long experience riding buses past Roslindale Square, they almost always lose half their passengers between FH and the Square. So if they took separate buses, the longer routes' passenger counts would plummet. The every-10-minute service along Belgrade Ave would probably drop to every 20 minutes or worse, for example. And would they really be much faster? They'd still have to deal with all the trucks, school buses, and cars on Washington St., not to mention all the buses that would still be picking up the crowds along the way.
edit, duplicate
.
Rozzie, West Rox, HP...
And Readville should be 1A.
How many of you on the Forest
How many of you on the Forest Hills side of the city are surprised that the 34/34E route is not in the top 15?
I was surprised, but as I look at the list, I suppose I shouldn't have been...so parochial, moi.
The 34
isn't on the list because 9 other routes share the most crowded section of the route. Total ridership on all buses running between Rozzie Square and Forest Hills is more than enough to justify extending the Orange Line to Roslindale. We need to get the new Mayor involved in some transit advocacy for Southwestern parts of the city.
What are they actually going to do about bunching?
The elimination of certain stops seems like a good move, so I'll give the thumbs up on that.
The T has talked about bunching before, but nothing has happened. Unless they explain what they're going to do, I'm assuming nothing will happen this time either.
The solution seems pretty simple; just send the first bus of the bunch "express" from a terminus until it gets back on schedule, or at least properly spaced. Passengers who want to get off an an intermediate stop should simply board bus #2 of the bunch.
Bunching is caused by overcrowding
And overcrowding is exacerbated by bunching, so it feeds back into itself.
A fuller bus takes longer for people to get on and off, as they negotiate narrow aisles blocked by other riders.
When a bus starts to fall behind, the number of waiting riders at each stop grows further, which means that when a bus does finally stop, it takes even longer to board everyone.
Skipping/passing-up stops can space buses out a bit better, and the T does do this, but it doesn't solve the real problem.
It may look like the buses are doing better because they arrive "on time" at the next terminus, but you've passed up a whole bunch of people to do it. That defeats the point of bus service, which is to serve people. And the crowds at the passed-up bus stops will grow to the point that when a follow-up bus does stop, it will be overwhelmed.
To make the best of a bad situation, when a trip gets dropped due to delays, it may be better to just send the buses out together as a convoy to handle the additional waiting riders.
Prevention is the best cure: speed up boarding and alighting with level boarding and curb extensions; faster fare payment (or use proof of payment); signal priority for more predictability at traffic lights; fewer, but better designed, stops.
Supposedly the key bus route improvement project does some of this. We shall see.
My suggestion still helps.
No, my suggestion still helps. There is no benefit at all to running the buses bunched. Although your solution is "better," it requires infrastructure that simply will not be built in my lifetime in Massachusetts.
Think about it:
If bus #2 is right behind the expressed bus #1, then the folks waiting at the intermediate stops will not have to wait much longer than they otherwise would by not expressing running bus #1. The fact that bus #1 passes them without stopping is probably frustrating, but ultimately of little consequence, if bus #2 is only a minute or two behind.
The folks further down the line (i.e., those past the "expressed" stop) benefit greatly, as the bus get to them sooner, and hopefully closer to schedule.
So why shouldn't they do this, at least in the short term?
Elaboration
I may have skipped a step and left you confused.
I am presuming that the reason bus #1 and bus #2 are leaving at the same time is because bus #1 was significantly delayed on its most recent trip to this terminus. Therefore, the subsequent trip is also delayed.
That means there is a longer-than-normal gap in frequency serving the bus stops. And during that longer gap, more riders arrived at the bus stops and started waiting. If there is enough delay, then the crowds at the bus stops could start to overwhelm a single bus, causing it to become delayed itself, and propagate the bunching backwards.
I have observed this in real life too, so I'm not just speaking theoretically here. Once in a while, I have found myself waiting for 30 minutes during rush hour at a key bus route stop, while being passed up by buses both full and half-empty. The full ones do not stop because they were at crush load. And the half-empty ones do not stop because they were trying to catch up to their schedule (presumably). The more frequently this happens, the more likely that the route is overloaded.
For example, Vancouver actually keeps track of pass-ups. They also have some of the busiest bus lines on the continent.
at long last proof
I've long said to my friends that "the 73 stops every 200 yards." Finally the map of that route shows I'm not far off, at least for the part of the run in between School and Mt. Auburn Sts. I'm glad they're eliminating stops. There's no reason to have them that closely spaced. (Which might have worked in the days when it was a trolley route, maybe?)
bus stop removal
No one has even mentioned one side effect of bus stop removal...more on-street parking! Where's MarkK when you need him?
former bus stop, future park(ing)?
No so fast, how do we know the former bus stops will turn into parking for cars? Maybe we'll get Hubway stands or private bike parking in there. Or maybe the former stops will continue to be officially off-limits for parking and fill up with the private cars of the well-connected. Maybe, as long as I'm dreaming, we'll get some bus-sized mini parks or food truck zones.
We don't need to dedicate
We don't need to dedicate more land to bike parking.
I've never had trouble finding a place to lock my bike. Except in Back Bay, but the problem there isn't a lack of land -- it's not enough posts to lock to, because of the multispace parking meters.
Net loss, of parking, I would think
If you look at the plans for route 1, they are expanding the length of many stops by removing 1-2 adjacent parking spots. I think this will probably lead to a net loss of spots, as the number of stops being removed altogether is small.
hey now...
....I was just being a wise-ass. You didn't have to take me seriously...
Check out the presentation
The Cambridge presentation on the #1 changes includes a table (slide 48) that shows a total of 69 parking spaces added, but only 7 removed.
Indeed, the exchange rate is about 4:1
Thanks ckd! The #1 bus route is slated to lose 15 stops gain 62 parking spaces net, a bit more than four parking spaces per bus stop on average. I'd still love to see some creative use of some of the reclaimed space.
It's good that they're
It's good that they're thinking about this.
But IMO the biggest causes of delays are fare collection, and badly programmed traffic lights. Removing some bus stops won't solve these problems.
For the first problem, they should make it easier to put a 7-day pass on a CharlieCard. And stop encouraging people to do the stupid value-loading procedure on the bus.
The whole point of the cash/ticket surcharge was to encourage people to use a faster means of payment. But what's the point, when they give the discount when you load money onto your card on the bus, which takes *longer* than just dropping in the cash?
I think they should eliminate the cash surcharge, and stop letting people add cash to cards on the bus. Using a CharlieCard should be its own reward.
And they should admit that the farebox coin and bill slot design is a failure. Next time, they should go with industry standard fareboxes, where you can dump in a handful of coins at once, and bills scroll by a window for the driver to accept rather than having a computer always reject them.
The traffic light problem would require some more effort and money. But that's easy -- the hard part is just getting the people in charge to make it a priority.