On urban bus routes, interior capacity is often cited as a pressing issue. A frequently proposed solution is to reconfigure the interior of transit vehicles to use more aisle-facing seats instead of forward-facing benches. In theory, aisle-facing seats use up less space, which provides more interior standing room and space to maneuver the carts, strollers, and various objects customers bring on board.

In theory. In practice. . .

View post on imgur.com

This picture was taken aboard an evening-peak NABI 60-BRT vehicle on the MAX route, operated by Transfort (Fort Collins, Colorado). In front of the rear wheel-well is a forward-facing pair of seats, with three aisle-facing seats occupying the wheelchair securement location. According to the website of the seat manufacturer, transverse (forward-facing) rows are manufactured to be between 35-37 inches in width, resulting in an individual seat width of 17 to 19 inches.

Notice how the feet of passengers sitting in aisle-facing seats protrude more into the aisle than the passenger in the transverse row.  The aisle-facing seats above the wheel-well have a gap behind them, as the wheel-well is wider than the length of a seat; but the seats in front of the transverse row are up against the interior sidewall. The customer in the transverse seat protrudes slightly into the aisle, perhaps an inch or two, and also has their foot rotated slightly outward into the aisle. In comparison, the foot of the customer in the aisle-facing seat protrudes further into the aisle.

If airline seating statistics are a good guide, the 95th percentile male transit passenger (the traditional seat design maximum [minimum is 5th percentile female]) will have a distance from the rear of their pelvis to the front of their knees of about 24 to 28 inches.

A passenger’s foot is. . . about a foot in length (you don’t say)! The seat back is about 2 to 3 inches at the top, sloping outward to be about 4-6 inches from the sidewall at the lumbar support. All together, a seat protruded 4 to 6 inches supports a customer with a pelvis-to-knee distance of 24 to 28 inches, with their foot protruding another 10 inches or so (subtracting dimensions counted twice), giving an overall protrusion of 38 to over 44 inches in length, more extreme than the 36 inches cited as the width of a transverse pair of seats.

In a transverse row, customers face the head of the person in front of them; in an aisle-facing seat, customers face standing passengers in the aisle. For obvious reasons, seated passengers don’t like to stare at the crotches of standing strangers; the eyes of a customer looking forward in an aisle seat is at roughly the same height as the crotch of the standing passenger. Customers also don’t like to have their foot stepped on or rolled over by other people, and expect the same consideration in return. As a result, standing passengers are placed in more uncomfortable situations, attempting to navigate the feet of seated passengers potentially with wheeled objects, resulting in passenger friction.

In crowded spaces such as a standing-room-only transit vehicle, people have a natural incentive to carve out personal space for themselves. In this example, the customer’s legs are bent slightly inward; on a more crowded vehicle, they may attempt to bend their legs outward to capture as much space as they can socially get away with.

By design, the foot of a customer in a transverse seat is underneath the seat in front of them. In an aisle-facing seat, empty space is created above their foot; in order to “use” this space a standee would have to awkwardly bow in to the seated passenger. While their feet could interlock, the desires of customers to maintain a personal space bubble as well as be able to stand up at their stop means this isn’t likely to occur between strangers.

This post isn’t to say that transverse seating doesn’t lack faults, the key one being legroom, but that solutions which involve converting more seats to aisle-facing should be given more scrutiny. In certain locations, primarily over wheel-wells and in wheelchair securement locations, aisle-facing seats are the only practical solution.

In comparison, a preferred alternative way to increase standing capacity is below. The most notable tradeoff is the reduction in seating capacity; aisle-facing seats can be substituted at roughly a 4:3 ratio, but removing transverse seats is a 2:1 reduction.


Photo by Ryan Flores/Flickr.

53 Replies to “The Space Fallacy of Aisle-Facing Seating”

  1. On the New Flyer buses here, as well as on the light rail cars, the aisle facing seats do seem to add a bit of space as the edge of the seat winds up being fairly far back from the aisle. The Gillig buses have much larger wheel wells, probably to accommodate the self-installing tire chain mechanism. In those, the aisle facing seats leave very little space in the aisle.

    However, the 2+1 seating definitely creates a lot of aisle space, and that seems to be standard on a number of systems outside the USA.

  2. The 2+1 seating can be found on Asian transit buses, however, longitudinal (aisle facing) seating is almost universal on Asian rapid transit trains. It’s worth noting that those trains are often wider than most buses, leaving more room to maneuver while maximizing seats per vehicle length. It’s prioritizing passenger flow over comfort since dwell time must be kept down to increase train capacity and many people don’t ride too long. Also, they have a higher tolerance of being in really close quarters.

    1. For link I’d like to see aisle facing seats in the platform level section and 2/1 seats in the raised car ends.

      For metro switch to aisle facing and a 2/1 layout as appropriate for the interior layout of each bus model. Some 3 door coaches like the rapid ride ones would be nice for high ridership in-city routes as well.

    2. I know someone who got on a very crowded trolleybus in St. Petersberg, and when he got home he had to explain to his wife how he got lipstick stains on the elbow area of his shirt.

      I’m not sure that the USA is ready for that type of close quarters.

    3. I’ll be honest here, I think that transit agencies need two fleets, one for base period service and one for the peak trippers with 1+1 seating and BIG standing areas. I know, I know, that means much more complicated parking at the base.

      It also means that occasionally, when there a cluster of “base period” buses happen to be at the shop, mid-day riders will not have the seat that they expect.

      I was a programmer for twenty five years, including doing big block transfers between memory locations, so it just does not impress me that when transit management says “Oooooohhh! That would be too haaaarrrddddd!” [Waving hands].

      Pardon my cynicism but something on the order of 30 percent of an urban agency’s fleet are trippers; how hard can it be to park them in their own rows at the base?

      1. It sounds like a good idea, but I think that the 1+1 seating fleet would be great off-peak for the short distance trips in various places as well.

      2. When I think of ‘cattle car’ bus, I think of what the Army used in Basic Training to move a lot of us around at once. It was Big Rig pulling a trailer that was standing room only.

        Load up! Load up! You! Get the f*** back! Push in! Think you’re too good to rub uglies with your buddy?!?! Get in on ’em! All y’all gonna be lonely tonight, make a friend now, save time later! Go on, push in! Push the f*** in! Make your buddy smile!*

        Cattle Car

        *When Link gets Seattle ‘crush loaded’ I get tempted to whip out my NCO voice and go to town. ;)

      3. I usually think of something along the lines of a Cuban Camel Bus:
        Or the Curitiba double articulated:

        TriMet has tried various experiments with various equipment, including borrowing an articulated bus or two from SoundTransit for a while (as it turns out, after the Crown-Ikarus fiasco, TriMet still isn’t ready for another round of articulateds). I’m not aware of any high aisle space experiment but I wouldn’t be surprised.

      4. Heck yes Matthew Johnson. I’m not very assertive, but the other day I was getting on the 43 at UWMC during rush hour, and had to yell at people to move back and make room. I would LOVE to hear more people directing folks to move back and make space at the front. Peer pressure FTW.

      5. Glenn,

        The profile of the Cuban “Camel” bus vaguely resembles a Bombardier Bi-Level Coach with no upper deck.

      6. Ever been on the Toronto subway? 1+1 seating and just freakin’ beautiful. The first time I rode one I missed my stop because I liked being in the train so much.


        Idea: Maybe make a larger portion of the seats foldaway? Having them all folded up by default – but available – would be a clever little nudge, sending a signal that passengers should stand by default. (Defaults are such a powerful choice architecture tool.) But there’s a seat there for when sitting is a must.

  3. On most buses the longitudinal seating is usually placed on top of the wheel wells because transverse seats wouldn’t allow for leg room at that location. In the top picture, if that is a 60′ foot bus, there likely is a double set of wheels on each side of the axle underneath those seats–which leaves less aisle space in the passenger compartment. If you look at the longitudinal seating on Metro buses that is above the double wheels you will notice that the back of the seats aren’t attached to the side wall of the bus. There is a gap between the back of the seat and the side wall of the bus. That space is necessary to create a seat that is comfortable to sit on. If the seats were flush to the side wall of the bus, the bottom of the seat would be so long that the seat would be uncomfortable for anyone shorter than 7′ tall. Longitudinal seating will work fine when it above a single wheel or when it is attached to the side of the bus.

    1. Yeah. Universalizing from the wheel-well experience, as this post primarily does, is pretty weak sauce.

      Yes, the wheel-well benches+legs leave as little aisle space as the 2+2 seating that proceeds it. That’s not the point. The point is that interior circulation, capacity, and dwell time suffer when such little aisle space is the norm throughout the bus.

      In the wheel well, reduced aisle space is inevitable. Everywhere else, more space can — and, on all but the longest-distance uni-purpose commuter-haulers, should — be made.

      Depending on the configuration of the vehicle, that can involve 2+1 or inward-facing or a combination of both.

      But the non-window-flush wheel-well example is wholly irrelevant to any of that.

      1. Don’t focus on the seats above the wheel well, but instead compare the transverse row on the left to the aisle-facing seats in front.

        Due to the built-in recline of the seats above the wheel well, the top portion is only about three inches from the back wall, which isn’t that much of a difference compared to the aisle-facing seat in front.

      2. Honestly, I don’t see it. And I think you may be applying too many of your own biases to the contrasting leg angles of three random people seen from afar in a blurry photograph. The woman in the forward-facing seat, for example, happens to have placed her feet together on the floor, yet her knee still appears to stick further into the aisle than any part of the side-sitters’ appendages behind her. (Compensating for the angle of the photo, I’d say it’s a difference of 3-4 inches at minimum.)

        I know that in most of Metro’s front-facing rows, I find myself needing to stick one of my knees and both of my feet well into the aisle in order to fit comfortably. And I’m of average height.

        The only certainty of your photo is that for a bus with empty seats (including the window seat next to the forward-sitter) and only a single standing passenger, it feels awfully crowded. Perhaps the wheelchair space should be standing/passing space at all times when not occupied by a disabled passenger. (Especially if the design of the flip-up seats causes them to stick out further than other inward-facing seats, a possibility you did not explore.)

        Or perhaps the bus just needs to be wider, period. New York never seems to suffer the problems you lament, not even at the wheel wells. A New York bus is only crowded when it, like, actually contains a crowd.

      3. Of course, a single photo from one trip demonstrating an anecdotal observation does not conclusive evidence make. Despite the limited number of photographed evidence, logical extensions, especially those rooted in geometry, such as the dimensions of a human body within a reasonable design range or the width and length of a seat, can still be made based off of the pictured observation and relevant data also available.

        Aisle seats do feel more spacious given that no seatback is adjacent to the center aisle, but this is only true when empty. When full, the body of a customer in the aisle seat of a transverse row protrudes only slightly beyond the seatback, but the body of a customer in an aisle-facing seat protrudes significantly beyond the edge of the seat. When the aisle-facing seats in your example are full, standees are unlikely to move any closer than where the yellow line is, even assuming that no surface height difference exists between the area underneath the seat and in the aisle.

      4. I think this is one of those situations where accumulated experience trumps napkin-theory calculations about perfectly regimented human body angle.

        There is quite simply a ton of room on those New York buses, up until the point when the bus truly exceeds its design capacity. The sideways seating allows for creative space negotiations as the available space contracts, in a way that “rows” never seem to regardless of their configuration. And this is especially important in New York, where bus routes may connect to multiple trunk subways along their journey, leading to high turnover and to sudden, major changes in load.

      5. On crowded buses a double row of standees seems to form in the front aisle facing seating area. Once the layout changes to front facing rows the standees are single file.

        For both trains and buses it certainly seems like there is more space with aisle facing seats both empty and full than with front/rear facing rows.

        Personally I’d like Link to be side facing in the platform level portion of the car and 2+1 in the raised portion.

        Streetcars should be open except for some single seats for the disabled in the low floor section (Portland uses a much better layout than we do).

        Buses should be a mix of aisle facing and 2+1 as appropriate. At least in the city and on busy suburban routes.

  4. Two more advantages of front-facing seats:

    They allow for more low handholds, as you can put a handle on the seatback of a front-facing single seat, but not across the lap of an in-facing seat.

    They also give you surfaces (seat in front of you, bus wall) against which you can lean a bag, which you also don’t get with in-facing seats: if I’m on a sound transit bus, this is usually what makes me pop my bag onto the overhead racks..

    The exception is that convenient side-facing area near the rear exits: I’ve managed to carry a folding chair from capitol hill to crown hill by sliding it into the space underneath the two forward-facing seats to my right.

  5. One of the things that struck me in Paris subway was that ALL seats flipped up. The unwritten rule during peak commute was that the sitters flipped their seat and EVERYBODY stood. Unless it was very very good excuse why you couldn’t. My ex-MIL discovered that being old was not an excuse.

    1. Seems like a lesson for Seattle. It would be really nice to have a way to accommodate people with disabilities that didn’t delay a bus up to 5min, inconveniencing all those other passengers so that a motorized scooter can be loaded on, rearrange people already in those seats on what could be a full bus, and half block the aisle only to get back off two stops later. Low-floor buses help but they don’t seem to be on the routes with the most disabled people. For some of these people it would literally be faster to motor their way down the sidewalk instead of getting on the bus. People on the bus miss connections because of this, when they would have had time to switch buses or walk to a Link station if they had known. I have a nightmare where people with disabilities all come out at once and force the buses to load and unload two of them at every stop, bringing the whole city to a grinding halt. It just sucks that taking transit is often a gamble.

      1. You better hope that you don’t end up in a wheel chair in the future and find out what it is like to be in that position and then have someone make similar comments like you did in your post.

      2. Right, cuz it’s so horrible to want better transit. I should be ashamed for bringing it up in a relevant forum.

      3. My daughter was in a wheelchair for 6 months and it got to where we didn’t go anywhere unless the Swift took us there since getting on and off in a wheelchair on the Swift was so easy and fast that we’d be locked in about as fast as the ramp can come up. On the RapidRide E it’s still very slow.

      4. Does Swift have the ramp at the second door, i.e. the one with the widest maneuvering space and nearest the passive-restraint spot?

        That set-up is incredibly common in Europe. It’s fucking baffling that we continue to force disabled passengers on a tight and invariably poorly-designed path between the driver compartment and the wheel well on services that pretend to be fast and scalable.

      5. Our last high-floor buses will be gone in the next year or two. It seems like it wouldn’t take much community pressure from a motivated disabled-rights activist group to get King County to find a way to fund universal conversion to rearward-facing passive restraints in that time. It’s the kind of one-time expense with long-term positive impacts that should be easy to sell.

        Suddenly, rolling on and off the bus is simply that – rolling on and rolling off. No straps. The driver doesn’t need to take off his seatbelt. And it’s that way every single time, on every single vehicle.

    2. Other than a few seats with a theater-style mechanism, positioned for wheelchairs to use next to the door, a cursory internet search does not reveal any subfeet with significant numbers of folding seats. I am finding that most new or recently refurbished subfleets have a 2+1 layout over some portion of the vehicle.

  6. I feel like this could quickly become a discussion for or against low-floor buses in general. From my understanding, low-floor buses offer faster load/unload times in exchange for reduced seating capacity. Most of that reduced seating capacity comes from the need to work around the wheel wells. And most of those buses (the New Flyer 60 footers) usually work around that by adding inward-facing seating above the wheel wells (in the middle) or with a “no seat, no luggage store” area (in the very front). Were it a high-floor bus, like Metro’s Gillig Phantoms, that wheel well is just another place for transverse seating (in the middle) or the extra-cushy priority seating (in the very front). Nevertheless, low-floors have pretty much won, so it’s kinda a moot point to argue for high-floors.

    One other consideration with transverse seating is that of getting off the bus. I tend to dislike sitting in the transverse seats if I’m not going far on a route because the inward facing seats always allow me access to the aisle which means I can get off the bus faster. If I’m in the window seat, sometimes I have to ask the person in the aisle seat to move so I get out. This is usually cumbersome, which slows people getting off the bus, increasing dwell times on fully-loaded buses.

    1. This, right here, is dead on. People moving in and out of the inner seats is a huge slow down, especially for city buses.

    2. Agreed, the declining number of seats available on a low-floor bus compared to a high-floor bus is a concern; space allocation decisions are more important with the comparatively reduced interior space to work around. As a comparison, the low-floor 60-foot bus in picture has 41 seats; the high-floor 40-foot Orion V buses operated by RTD in Denver have 43 seats. 20 more feet of bus, but two fewer seats. It may seem trivial, even beneficial, to remove up to two seats in the elevated compartment to increase legroom, eight seats for four interior bike racks, three to four for a third door, or six for passive wheelchair restrains with theater-style mechanisms, but considered in totality a significant loss in seating capacity is the result. Whether additional legroom for the remaining seated customers, interior bike racks, the mobility advantages of a third door, or the freedom of a passive wheelchair restraint are net beneficial is up to debate; but the inherent tradeoffs must be acknowledged.

      As you hinted at, people tend to self-select their location choice to gain certain advantages. Customers travelling together may try for adjacent seats if possible, the solo customer going a long distance may try for a window seat to have a surface to rest their head against, and the short-distance customer such as yourself may try for an aisle seat or stand if no seats are easily reachable. My assumption would be that any additional friction caused by a greater number of transverse seats would be negligible, and possibly offset by the decrease in friction from fewer feet noticeably protruding into the aisle.

      1. Why couldn’t they have the front door be low-floor for wheel chairs, bags etc.. and then have stairs up to a high floor for the back two doors? Does anyone make anything like this. We sort of do it for the Streetcar.

  7. I can’t speak for BRT’s other than the ones that I’ve used. KC Metro uses New Flyer low floor buses and the majority of seats on the coach are side-facing. My comment is that these side-facing seats are uncomfortable in the extreme as opposed to the few front-facing seats.

    1. I personally think side facing seats are the least desirable with the back row following close behind. our bodies don’t seem to be designed to keep us upright if the forces are coming from the side eg. stopping, starting making it difficult to ride on a bus that brakes hard or accelerates hard.

      I only sit in a side facing seat if there’s no other option. Except in the Swift where I take the side facing seat behind the driver and turn around so I’m facing the back of the bus and lean against the padded wheelchair stop.

  8. I’ve read that Washington State highway lanes are more narrow than in any other state.

    Do our buses have smaller widths than elsewhere because of this?

    1. Do you have a citation for your assertion that WA has narrower highway lanes? Our freeway lanes are mostly standard width, but there are exceptions like the pre-rebuild 520. After R8A phase 3, the I-90 bridges will have narrow lanes.

      How much extra do you think it would cost to order buses that are narrower than everywhere else?

    2. Where is that citation, John? I’d love to read it. Was it a comprehensive study by a university or FHWA/USDOT? I’m sure a few other states would be on that list too.

      Standard freeway lanes are 12 ft in width per WSDOT design manual. This even aligns with the AASHTO green book. However, you may find 11.5 ft and 11 ft lanes (with approved FHWA design deviations) on some segments of freeway. Nothing you won’t find on I-680/580 in Alameda County, California.

  9. Aisle-facing seats are proven to pack far more people in than tranverse seats on *trains*. Which is why nearly all subway cars have aisle-facing seats (or, indeed, benches). But this may not translate to buses, for various reasons…

    The wheel well bumps and the narrow vehicles certainly seem to be an issue. With trains, aisle-facing seats usually leave a pretty sizeable aisle; not so much on the buses you photograph.

    1. I was on some Boston Green Line low floor LRVs a few weeks ago that are pretty constricted in their articulated section. There is longitudinal seating there, but four of the eight or ten seats have a very short seat length. Knees are sticking out into the aisle more than feet.

      1. Yeah, Breda sucks at everything. Those trains are so dysfunctional that most of the time they still get pulled behind Kinkis 15 years older than them.

        But thanks to an open floor plan everywhere outside the articulation — high-floor as well as low-floor sections — you can (and every rush hour, do) cram 4x as many people on the cars as have ever been seen on our supposedly “packed” Link trains.

    2. Aisle-facing seats on buses are often quite horrible things ergonomically speaking. Buses jostle far more than trains, sometimes quite violently, especially in the forward and backward directions for which aisle-facing seats offer next to zero support and scarce objects to grab on to. Buses also operate on much steeper grades, so you’re constantly smooshing into the person next to you, which is usually unenjoyable for at least one involved party. Not that light rail passengers are into free love, mind you, but it’s a much more pleasant way to travel even if you’re still packed in like sardines.

  10. A segment of the population gets motion sickness easily when riding sideways. I don’t know what that percentage is, but I’ve talked to many people who have this problem.

    Some additional factors seem to be the number of starts and stops, the amount of turning on the vehicle, and the length of the trip.

    Has anyone ever seen a study on buses and motion sickness?

  11. Key take-away: forward-facing seats save room because it’s like spooning, but without the physical bubble issues? :P

Comments are closed.