.
.
.
.
.
.
.
.
.
King County Metro Route 60 has received at least two large investments from 2014 Seattle Proposition 1 funds that extend span of 30-minute-or-better headway all the way to 7 am to 11 pm seven days a week and, as of this September, now provide 15-minute headway from 7 am to 7 pm on weekdays.
Route 107 has also seen good investment from Metro since it was rolled out last fall, with 30-minute headway all the way from 7 am to at least 9 pm, seven days a week, and a perfectly-placed layover spot that allows it to round the corner to the southbound bus stop across from Beacon Hill Station.
Both serve the corridor on 15th Ave S from Beacon Hill Station down to Georgetown, with whichever bus comes along first picking up the bulk of bus riders at any of the stops in that corridor.
However, the two routes are not timed to provide coordinated headway on the corridor, producing a painfully wasteful service pattern.
During weekday evenings, routes 60 and 107 pass Beacon Hill Station southbound at almost the same time. Northbound, route 107 is scheduled to arrive at Beacon Hill Station 11-14 minutes before route 60.
On weekend afternoons, the two routes are scheduled to pass by the station southbound roughly two to three minutes apart. Northbound, they are scheduled to arrive roughly eight to eleven minutes apart.
With cleverly-coordinated timing during evenings and weekends, routes 60 and 107 could provide roughly 15-minute-or-better headway on 15th Ave S between Beacon Hill Station and Georgetown from 7 am to 9 pm or later, seven days a week.
But even if that is easier said than done, having at least 10 minutes between each bus would mean that 2 out of 3 trains going each way would have a bus to which to transfer to head down 15 Ave S, and riders headed to the station from 15th Ave S would be able to catch the next train 2/3rds of the time, the following train 1/3rd of the time, for most of the evening and weekends.
Back-to-back routes are not entirely without usefulness. It gives you redundancy, in the sense that if one route gets delayed, you can still catch the other one. It also means you can plan for the first bus in the bunch with much less wait time at the bus stop if the presence of the second bus means you don’t have to be too careful about missing the first one.
Yeah, sure, but it is still better to have buses equally spaced. In an ideal world you have buses running every five minutes or so — very little waiting, while minimizing bus bunching. In this case, you have the possibility of bus bunching, along with some very long wait times. You still have to time it, which might end up costing you a huge amount of time if you need to get somewhere at a specific time, or are reliant on another transfer (to a similarly infrequent bus).
At the same time, a rid on the bus might be delayed by the other bus. Bus bunching in this case is worse than when the same route gets bunched. Two buses serving the same route can leap frog each other, with the crowded bus only making drop offs. But in this case, the two buses have to serve each bus stop, which means that one might have to wait while the other is delayed (by, say, someone in a wheel chair). In this case, it really is the worst of both worlds (low headways and the possibility of delays caused by other buses).
Bus bunching is built into the current 60 and 107 schedules, particularly in the evenings. During the middays there are 6 buses per hour along 15th Ave S (4 x rt 60, 2x rt 107); but they aren’t evenly spaced and the 107s are usually running headlights-to-tail lights with the 60s. At night it’s even worse when the route 60 drops to 30 minute headways. The 60 and 107 are often scheduled to leave BHS within 2-3 minutes of each other.
Are there lots of wheelchair riders along this route? Then why don’t we make it a streetcar to improve the rider experience?
I guess it might be tough to put rails across the Albro Place bridge.
I’m not opposed to the concept of streetcars and I sometimes defend them, but this is not a corridor that would ever support a streetcar.
Wrong fish took the bait. Oh well.
I was wondering if you were serious…
In theory, evenly spaced buses minimizes the wait time of somebody showing up at the bus stop at a random time, without looking at the schedule. I actually sat down and proved it once for the general case of n buses. Some interesting algebra and vector manipulation involved.
That said, whether a schedule shift is best in this case depends on what impact it has on other transfers elsewhere in the routes. And there still is something to be said for knowing that when you’re schedule is tight, that you have two routes to choose from, only one of which, has to actually show up.
Waiting 10 minutes average is better than waiting 28 minutes if you just missed the bus. The 39 and 75 used to have the same problem, where if you missed the 30 because of freeway traffic or bus overcrowding, the 75 would come 2-5 minutes later (because all routes were unreliable). But that meant there were only two times per hour you could make the trip. The same thing happened with the eastbound 43, which came a few minutes before the 14 and cannibalize d its riders, mainly because people didn’t trust the 14 would actually show up 3 minutes later: sometimes it would be ten .minutes late or wouldn’t show, and the next bus would be another 43 (so you waited in vain). If the routes are scheduled 15 minutes apart, then it’s more likely one of them will come soon because it’s unlikely they’ll have the SAME amount of delay at the same time, unless they’re stuck in the common segment, in which case there’s nothing you can do e cost wish for transit lanes.
>> That said, whether a schedule shift is best in this case depends on what impact it has on other transfers elsewhere in the routes.
Yes, in almost all cases it improves them. The only reason it would be better is if the connection happened to be timed just right, but in that case, all it takes is for one of the buses to be late to completely screw it up.
It isn’t just a matter or randomly taking a bus, it is also a case of trying to be somewhere at a particular time. Let’s say I want to be somewhere at 6:00 sharp. The buses arrive at my destination every half hour, either on the hour, or 5 minutes later. If I take the bus that is supposed to arrive at 6:00, I am late most of the time. so that means I have to take the bus that arrives at 5:35. If the buses run every 15 minutes, then I aim for 5:45, a ten minute savings. Now imagine I am trying to catch another bus that crosses at 6:00 as well. Same story. If I aim for a perfect transfer, I will miss it, and have to wait a half hour most of the time. So I aim for the previous bus, which means the 5:35 (if the buses are close to each other) or the 5:45 (if they aren’t). Going the other way, you could try and time the system just right (bus arrives at 6:00, other buses arrive at 6:00 and 6:05) but again, that is pushing your luck. If the first bus arrives five minutes late, you are in trouble (and have to wait 25 minutes). You could try and time it with the train — which has less variance with its timing — but now you are trying to figure out the time it takes to get out of the station and to the bus stop. The train also runs quite often, meaning at the end of the day, it doesn’t matter much — you are back to worrying about when you will arrive at your destination. In short, no matter how you draw it up, with a bus line like this (which is by no means a Link shuttle) equal spacing will work out better for most trips.
Now, what these buses do on the rest of the route might be a different story, and be an argument for keeping the current schedule. As it turns out, I think these buses operate independently (they just turn around) making the proposed change a very easy one to implement.
“whether a schedule shift is best in this case depends on what impact it has on other transfers elsewhere in the routes.”
Yes. In the past the 101 and 106 were coordinated to provide 15-minute service from downtown but not in the other direction. And the Snoqualmie Valley Shuttle is timed with the 224 at Duvall but not on the other end with the 208 in Snoqualmie. So the question becomes, is the 60 or 107 timed with anything more important than Beacon Hill station? I don’t know the 107’s Renton commitments. But the 60 doesn’t seem timed with anything in Broadway, South Park, or Westwood Village. In Broadway the only northbound transfer is the 49 to the U-District, but for that you’re better off transfering to Link at Beacon Hill. Transfering southbound from Beacon to Rainier is probably nonexistent or marginal, and again Link is better for some destinations. In Westwood Village the 60 stops is on the opposite side of the shopping center from all the other routes, which makes transfers both inconvenient and easy to miss, so that doesn’t seem to be a priority. South Park I know little about, but are the northbound routes at the same stop and is there any sense in the 60/132 schedules? In Georgetown the 60 overlaps with the 124 so that might be an issue since Metro recently reorganized routes down there to make Georgetown more accessible. But on the whole, Beacon Hill station seems like a very good place to coordinate routes.
During evening and weekends, I can almost set my watch by route 60. It doesn’t have to go through downtown. It is really dependable. It is my route of choice, except when I see route 132 coming, and cross the street to take it instead going north. But to go south, route 132 is totally random, and the stops are super-sketchy, so I take the train to BHS, and know when the 60 will arrive.
Route 107 is equally dependable going south, as it lays over around the corner from the station. Since I am headed to South Park, I wait the extra couple minutes after the 107’s break ends and it comes around the corner, picking up all the other riders except a couple waiting for the next 36, and route 60 shows up within a minute of its schedule. I don’t know how reliable route 107 is northbound, as I rarely have a reason to ride it.
The 107 is probably reliable as it’s all on back roads with little traffic except in downtown Renton (which is a reliability black hole).
That’s fine in the southbound direction. Northbound, whatever delays get encountered in Renton will get propagated throughout the entire route.
I’ve got some half-worked out suggestions for scenarios that would improve service along 15th Ave S. that I’ll present in conceptual form.
Concept One: disconnect the 107 from the 148 and adjust the 107 schedule to complement the 60 schedule instead of mimic the 60 schedule. Under this concept the 60 and 107 schedules would be coordinated with the Link schedules to provide optimal transfer opportunities at BHS. When Link is running on 10 minute headways the 60 and 107 schedules could be coordinated so that 4 of the 6 hourly Link trains would be scheduled to provide an almost instant transfer to either the 60 or 107. Under this concept the 148 could be connected to the 105 at the Renton TC.
Concept Two: this would provide a bus transfer at BHS every 10 minutes. Take Concept One and revive the old 38 that ran from Mt. Baker Station to BHS via McClellan but extend it on both ends. The neo-38 route would start at the 14’s Mt. Baker tail and serve Mt. Baker Station, McClellan St., Beacon Hill Station, 15th Ave S, Georgetown. The neo-38, 60 and 107 schedules would be planned to provide service every 10 minutes between BHS and Georgetown. The 14 would drop its forked tail routing and those service hours would be re-allocated to pay for some of the neo-38’s hours.
The neo-38 would require one bus and dropping the 14’s forked tail would save one bus so the neo-38 plan would be revenue neutral for Metro
Good thinking out of the box on the 14′ s tail. It turns the tail into part of a grid route.
Here’s the 15th Avenue South Service Improvement Plan (Proposed)
1. The 107 drops the connection to the 148 and runs as a Renton Transit Center to Beacon Hill Station route with an extension to the legacy tail of Route 14. (6 buses would be needed but 1 bus could be removed from the 1/14 schedule for a net of 5 buses)
2. The 148 would be connected to the 105. Five buses would be needed for this combination but it would be a very long route because the 105 is currently a loop route beginning and ending at Renton TC. To keep schedule reliability and ensure the existing timed connections at South Renton P&R with the 101, Metro could create a layover space in the Renton Highlands or extend the 105 to the Kennydale P&R for connections to the 566.
These 2 changes would be “service hours neutral” because the current 107+148 uses 8 buses and the current 105 uses 2 buses.
The 60 schedule would remain the same as it is now but the 107 could be adjusted to provide the more coverage without duplication, particularly during the evening hours.
I still don’t think the #14 tail needs any bus service, period. It’s close enough to Mt. Baker Station to walk, down a pleasent, tree lined street, and, most of the time, faster to walk than to wait for the bus, especially if one is trying to get to Link, since you can take the ped bridge and avoid the transit center. Also, the number of people that actually live by the 14 tail is small, and the ridership there, negligible.
Agree and agree, but I think the legacy tail on the 14 is served mostly because Mt. Baker TC doesn’t have enough space for the 14 to layover there. The buses take their long layovers on the legacy tail and pass thru the TC only to pick up passengers.
I’ve always liked the Metro plan idea for the Route 14 tail, which is to extend Route 27 to Mt Baker Station via Lake Park Drive as the replacement. It seems so silly to have two almost empty, dead-end routes (Routes 14 and 27) about a half-mile or two minutes from each other, when extending one would offer residents in two areas with two directions to ride directly.
Richard, ‘fraid you’re right about streetcar all the way across from West Seattle. Though future LINK transfer to Beacon Hill and MLK might work, making real progress possible.
Very likely LINK fleet has curve radius small enough to be switched to Renton Avenue at Othello, and with reserved lanes, negotiate Rainier Beach, and then head out along the lake to Renton. Pretty much like former 107 used to do, except would pass airport, not go through Lakeridge.
Present wheelchair securements pretty primitive. Could be a lot faster too. So equally valuable fallback. Continue the Rainier Avenue line through Rainier Beach, and then down the lakeshore. Same substation could probably handle Prentiss Street into Renton via MLK.
Gothenburg already dual-modes triple section artics with streetcars. But where there’s a will there’s a way. As long as the millionaire didn’t leave it all to his cat. And also:
https://www.youtube.com/watch?v=Y5uZXXWKvxo
OK, it was a mammal, but definitely the wrong one. Though positive result was that Ishmael survived to found Sea-Shepherd. But re-make where Tom Cruise nails his credit card to the mast is now floating in that huge mop of plastic in the Pacific, VISA and all.
Mark
Minor correction: the route 107 was not rolled out last fall — it was extended beyond Rainier Beach.
Also worth asking if it’s worth having the 107 do its Georgetown detour in the first place, if it’s coming at the same time as the 60, which is already going to Georgetown anyway, from the same place.
Maybe instead of updating the schedules, it might make more sense to just straighten out the 107 route and speed up the rest of the trips. This move, of course, might not be feasible if people use the 107 to access Georgetown from Renton/Ranier Beach, rather than from Beacon Hill. Does anyone actually use the 107 this way?
Okay, I’ll be the one to ask, what IS “headway”? It seems important somehow, but after this article I’m still not sure what it is.
From the STB Definitions page:
In other words, it is the frequency of service.