In Denmark, the term højresvingsulykke (right turn accident) is common, meaning when a car, typically a truck, turns right and hits a bicycle. Lots of people are killed from this every year. One theory is that it is primarily foreign truck drivers who are not accustomed to the high number of cyclists in Copenhagen.
When I initially saw the title, I was expecting it to be a mistake and actually say "Why UPS Trucks Don't Turn Right". I guess the problem is less severe in the rest of the world.
When I cycle, I never pass a car on the right at a crossroads. Even if they are not signaling a right turn. Regardless of who has the right of way there, I am going to lose a vs. a car.
Indeed. The right of way is indeed cold comfort when you're DEAD. Of course, some times you'll be approaching a crossroads and a car will be approaching it faster and pass you just in time to run you down. :)
This is why I have a bell on my bicycle, and ring it any time I have a shadow of an excuse. (I recommend spending a few dollars extra and getting a nice bell that you like the sound of.)
This is a situation where bikes don't necessarily have right-of-way under the law. If there's a bike lane, then yes - and in that case there really should be traffic markings to make it clear to motorists that the right turn lane is on the other side of the lane.
But if not, then bikes are using the same lane as every other vehicle on the road. So when cyclists pass waiting cars on the right at a stoplight, it's just as dangerous and illegal as if a motorcyclist or scooter rider were to do it.
That depends on where you are. In Germany it is perfectly legal for a cyclist to pass waiting cars at a stoplight on the right if there is enough space, because it is safer for them to wait a the front of the line than somewhere hidden between cars. Especially if they can get to a marked waiting area like this: http://commons.wikimedia.org/wiki/File:Radaufstellstreifen_%...
Seems like cyclists dangerously undertaking (passing on the incorrect side) causes these problems. Much to the annoyance of cyclists, smart motorists would position the vehicle to prevent undertaking maneuvers when approaching a turn. For cyclists intending to go straight, the cyclists needs to take the lane, positioning themselves so they can't be overtaken by vehicles intending to make a turn.
In all cases, motorists needs to be aware of the more vulnerable road users and cyclists should take more precautions when sharing the road with 2+ ton vehicles.
From a cycling commuter for the past 5 years, 37km one way, in the Tokyo Japan area.
Some US cities have that, too, but not all of them. In the ones that don't mark spaces like that, though, there isn't really space for cyclists at the front and even pedestrians aren't particularly safe from motorists looking* to make a right turn at a red light.
* On further thought, maybe that choice of word is inappropriate.
as a cyclist in the US there's a very common scenario here, too - turning right, looking left, not looking into the turn and bam, cyclist at high risk.
TO CLARIFY - the situation i am most referring to is a cross street, where the car is perpendicular to your path and would be coming from your left as a cyclist.
At least in California, the car is supposed to take the bike lane rather than turning over the top of it. In this way the bike only ever passes the car on the left, and motorists (provided they look when taking the lane) are less likely to hit a cyclist if they fail to check their right-hand mirror again.
He might not have been downvoted if he hadn't said it, but it's not like it's a farfetched idea. In Toronto Mayor Ford practically made it a campaign promise to screw over cyclists, going out of his way (and incurring great cost) in order to rip out recently installed bicycle lanes. So it's definitely a political issue, and while there might not be actual bona fide lobbyists there are definitely voters who are anti-road-cycling.
From the sounds of it, an optional bike lane would be a marked lane that only bicycles are allowed to use, but where bicyclists can also legally use the other lanes on the same road.
Imagine the bike lane is on the right. Now if cars are moving slowly you as a cyclist will be passing them on their right ("undertaking"). This is unsafe near the places where cars turn right (across the bike lane).
1. A bicyclist does not need to stay to the right of the road when approaching a place where right turns are allowed (i.e. can leave the bike lane).
2. Vehicles are required to merge into the bike lane before making a right turn, otherwise it's consider making a turn without being in the rightmost lane.
Not to say I've ever seen a single car in the bay area correctly merge to the bike lane before turning. Any case where a bicyclist is hit by a car making a right turn should be considered "failure to complete a passing maneuver."
A complication is when a car (or bus at bus stop, or whatever) is parked in the bike lane, or immediately on the curb lane, just before the right turn lane. In that case, it becomes a lot harder to get into the bike lane before making a right turn, and in SF, is a fairly common situation.
If you're just worried about cars, it's fairly safe to make the right turn lane from that non-rightmost lane, since the other car is blocking any other car from approaching there. However, bikes (and sometimes motorcyclists, but rarely are they that suicidal) are going to come up, so it's still prudent to both get into the far-right lane and check on the right. That's the most common "cars turning right from non-rightmost lane" situation I've seen, though.
My biggest question about California road rules: what kind of vehicle is a homeless guy with a shopping car full of cans/etc.? Is he a pedestrian (similar to a pedestrian pushing a stroller), or is he a vehicle like a bicycle? What if he's in a lane of traffic, going down the road at 1-2mph (which is reasonable, since he would take up the entire sidewalk otherwise).
A bad idea perhaps, but legally required, at least in Denmark. Bikes are suppose to stay as fare right on the road as possible at all time. This means that traffic aren't mixed, which is generally a good idea.
Isn't that pretty much why you always cycle with the direction of traffic? As a runner, I've been told to run against the flow for safety but almost always run into these types of situations (which is why I mostly go out of my way to run behind the cars stopped at intersections.)
This is a type of accident that is particular to countries that have bicycle lanes. Cities like for example Berlin have closed a large number of cycle lanes (where they are routed to the right of regular traffic) because of that reason, and replaced them with shared lanes.
That's a sensible policy. Here in Jakarta lanes are mostly for decorative purposes only, but the roads are shared by cars, trucks, autorickshaws, motorcycles, and various forms of human-powered cards (either pushed by a pedestrian or with a mechanism based on a bicycle). It is amazingly chaotic.
But one of the first things you learn is not only to be pedantic about signalling when turning but that you need to claim space because otherwise you are asking for an accident and if you are involved in one as a car driver, it doesn't matter whose fault it is: the motorcyclist/bicyclist loses on the physics and the car driver loses in front of the police (that's one way btw that police corruption protects the poor here).
The space-claiming is key. You do not want to be passed on the left while making a left-hand turn (they drive here the same side as in the UK). That has happened to me a couple of times and it is a major reason I do not listen to the radio while I am driving (with motorcylces that way at least I can hear them). Fortunately I haven't hit anyone yet (knocking on wood).
In most, if not all, cities in Canada, this is why we're told to always check our right-side blind spot before we turn right. Don't know how it is in Denmark. A quick Google search tells me that Denmark doesn't have such a law. Not sure why they don't, it makes total sense.
Not looking over your shoulder when turning will give you an instant fail in the driving test in Germany. In the US most states' drivers manuals recommend looking over your shoulder but most people don't do it. I tend to blame it on the fact that there exists no mandatory driver education and driving tests are almost impossible to fail.
When I paid several thousands of Euros for mandatory driving school in Germany I was very annoyed about the high price tag, but seeing on a daily basis how safety and efficiency of traffic is compromised when no system to educate drivers is not in place, I am glad I had a chance to go through it.
Each US state has its own licensing laws and you may have issues with the particular laws in your state, but it's certainly wrong to say there's no mandatory driver's ed in the US. California, for example, requires 30 hours of classroom education and 6 hours of behind-the-wheel instruction from a state-licensed professional driving school.
I have a CA license and my US driving school track record consists of a total of 5 hours of "classroom education" (video watching) and passing a driving test despite running a stop sign in a residential area (my German license didn't count, had to start from nil)
It's reversed in the US. Lots of accidents occur when people make a left hand turn and don't see a motorcycle/bicycle going straight in the oncoming lane.
Indeed. A big problem in DC is cars making U-Turns (i.e. a hard left turn) and cutting across the bike lanes that run along the center of Pennsylvania Ave.
I think this can be generalized to: Turning and/or crossing lanes results in more potentially dangerous situations than traveling within a traffic lane.
Cyclist deaths get a lot more attention that purely automobile deaths, because they are so rare, which makes them seem vastly more common than they are.
In fact, in 2013 there were 14 cyclist deaths in London. 6 of those were in a two-week period in November. That was shocking enough that there was political attention and increased police enforcement around intersections. I'm sure also increased media coverage.
9 of those deaths were from lorries, as you said. I'm sure the majority are also at intersections, as you said, since that's where most cyclists get hit by motor vehicles.
You have a point, however saying that a cyclist dies every week in London 'because they are not wearing helmets' is equally wide off the mark and erroneous.
When I initially saw your post, I thought that Denmark was driving on the left... but then I realized that much less countries then I initially thought are driving on the left side.
I doubt that lots of bicyclists get killed this way every day in Denmark. Sure, it's a risk and there're certainly accidents and deaths, but many and every day in a relatively small country?
The streets in the picture were at least designed by someone with half a clue, and most drivers would hesitate to turn right at speed here.
This is what this normally looks like (theres no collision here because as a cyclist you constantly monitor other lanes and slow down until its clear the idiots have noticed you, or just switch into the full lane):
I have a friend that is a UPS driver and he always rolls his eyes when somebody mentions this to him. UPS trucks definitely turn left. Sure the delivery order generally tries to reduce left hand turns, but on the routes that he drove he saw very little difference in the route when comparing them before this 'change' and after it. The biggest cost savings were done by reducing the driver and truck counts and paying the remaining drivers overtime to deliver more packages.
Actually if you read the entire article it addresses the fact that while UPS finds most left turns inefficient there are some instances where their routing software recommends a left turn.
Cutting out left turn makes a bigger difference in cities where turns are more frequent and often require waiting due to oncoming traffic. It saves fuel for UPS, else and for everyone who isn't waiting in a lane blocked by a UPS truck that is waiting to turn left. Now if we can just get others to stop turning left (when it requires blocking a lane). Since not all routes are in cities with dense traffic, not all routes will get the same benefit.
I always get a mental jolt when I see a UPS truck turn left because of this "rumor" or perception or whatever it is. I certainly see it, and remember it when it happens because of the idea that they never do turn left.
Interesting article! I remember watching the mythbusters episode on it.
It all depends on what's more efficient routing wise. Time is money. Not all left turns are bad...
Id very much like to know why they don't send a text+email prior to a physical delivery so I can confirm whether I will be home to take delivery and potentially offer a reroute address or simply say do not attempt, rather than the current approach of making three unsuccessful attempts.
My old FedEx guy used to do this. It was great. Then they took away their phones as a cost-saving measure... Now I never order anything to ship via FedEx. Hope they're saving lots of money.
I recently consulted with a logistics company. Turns out reroutes and missed orders are expensive. Really expensive for the company. Since the company also used owner drivers, the cost of the reroutes are put on the drivers themselves. Naturally, the incentive is not to reroute.
Actually, after working with them for a bit, I am a lot more sympathetic to posties who just leave stuff at the door.
At least for FedEx, rerouting individual packages costs money but vacation holds are free.
Since I've never encountered a day where I'm expecting two packages to be delivered by my driver and I'll somehow be around to receive one but not the other, I've never seen much reason to go with the non-free option.
You can "Hold for Will Call" (Request packages to be held for pickup at a UPS facility) for free -- but the usefulness of that will depend on the hours and location of your local UPS facility.
You can also sign for the package online or provide the driver with "leave with..." instructions.
If you have the my UPS app they will send you a push notification to your phone. It is possible to reroute it with the app, but I've never needed to; since the UPS driver will most likely just leave it.
In my experience, only freight companies call ahead.
I'm working on this exact issue with a delivery company currently. The biggest question they have is "would you pay for it?". Would you pay extra to receive a message to advise of delivery time, and would you pay extra to have that delivery delivered at another time or elsewhere?
Whilst missed deliveries are costly they obviously don't feel that is justification enough to develop a better solution.
My own personal opinion is no, I wouldn't pay for this service. However, maybe the costs can be driven down by simplifying the solution. Instead of 'deliver wherever the customer wants,' only present the customer three options: 'Receive at home,' 'Receive at work,' 'Don't deliver, I'm out of town'. This solution perhaps reduces the complexity in customizing delivery location (fewer possibilities), pushes down delivery costs (fewer missed drop-offs), and increases customer satisfaction (receive package with less hassle).
Some of the consumer delivery firms that Amazon use in the UK have started texting the night before (or early on the day) with the option to either delay delivery or to deliver to alternative -- but the alternate delivery address must be in the same postcode (between 1 and ~20 houses) so no re-routing is required.
I imagine it'd depend on the density of deliveries. Once the truck is loaded and navigating a town, the marginal cost of a re-delivery is probably negligible compared to the logistical problem of recomputing optimalish routes and attendant schedules on the fly.
Whenever I get annoyed at labor unions and think they're mostly inefficient rent-seekers, make work, etc. (which they tend to be, in declining industries), the counter-example of UPS comes to mind -- they're unionized, but in a stable or growing industry, and seem to have both fair (not too high, not too low) wages and great performance.
Apropos left turns, former FBI director J. Edgar Hoover was involved in a left-turn accident early in his career, after which he ordered that his drivers never turn left when he was in the car.
Quote: "A small army of agents spent days planning details of the trip. The supervisor drove the car with Hoover and Tolson. Hoover insisted on riding in the back seat, on the right side. He’d once been injured in a car wreck while sitting in the left rear seat, and he refused to sit there again. Also, the accident occurred during a left turn, and Hoover no longer allowed his drivers to make left turns. This complicated the route to Austin in those days before interstate highways."
For people who don't live in North America: If you're at a stop, you follow a "free right" rule where you're allowed to go right even if the traffic lights are red, but as long as there aren't pedestrians crossing.
I can see the free-right policy also being a major factor in this as well but I don't see it mentioned.
What most people in North America seem to conveniently forget about this rule: You have to come to a stop first.
Another piece of trivia about free right: In all of New York the free right is allowed as long as no "No turn on red" is posted. In New York City, however, the free right is prohibited unless a "right turn on red" sign is posted.
In Germany the NYC-style free right exists, with a small sign showing a green arrow on black background indicates that free right is allowed.
You really need a data driven approach in this. Gather statistics of actual vehicles going from point A to point B so that you know the best way to do it in the future.
I'm running a routing optimization service (http://fleetnavi.com if you are curious, but it's only in Bulgarian for now, although Chrome translate works really well on the site). The routes a mapping service like google maps will give you are not that great really, in many cases the data needed to do a perfect route are just not available. We aren't using the stats from routes in the past yet, but it's a next step that would clearly be beneficial.
Yes you do get some uneven wear, but I'm sure UPS's fleet maintenance keeps track of that too. For the tires, you can rotate them on a regular schedule to reduce repeated wear patterns and spread it more evenly on all of the tires before they're replaced.
In the movie 'Cop Land' Ray Liotta's character (a cop) repeatedly uses the phrase "Don't fight, go right" to refer to a tactic fugitives use: avoiding left hand turns. Apparently, if you know that you can chase better.
I think it was meant as a double entendre too - referring to one's moral compass.
I can imagine that in a typical American city, but I wonder how this would work in a typical European city. Where you don't usually have so nice even blocks like in the US.
I could get to my parents w/o a left turn. So I don't think it is impossible to avoid left turns in a European city but I think it is much, much harder to do so.
Aside from route optimization, something else that the tracking devices may allow is recording acceleration, braking and instant fuel consumption. This in turn makes possible to train and reward drivers for more efficient driving.
It's a big selling point for systems sold to trucking and public transportation companies.
I'm reminded of the one time I drove through New Jersey. Major arterials are all set up like interstates and left turns aren't an option. Usually you have to go further than your destination and take a right to get on an overpass that spits you out in the opposite direction on the arterial. It's very frustrating if you manage to get lost.
It's hard to imagine that a high number of intersections are so bad for left turns that a three right turn detour is warranted. I've seen some left turn lanes get severely backed up during rush hour on the busiest of streets, but as a general rule I'd expect the left turn would win. I'd be curious to see more data on this.
Depends on a lot of factors and as mentioned by ceejayoz, the routing is likely a lot more complicated than that, taking into account different trucks operating in the same areas, but I know of quite a few intersections (stop-light controlled rather than stop sign controlled) in my neighborhood where I prefer multiple rights to a left because right on red is allowed (if no oncoming traffic) and you can often make the 3 right turns much faster than waiting for the one left turn at a light.
In most situations I'd imagine the routing systems can avoid the three-right detours just as much as avoiding left turns. Another truck can likely handle the other road on its route.
From TFA - it's not just the right turns that saved them that much fuel though -
> As of 2012, the right turn rule combined with other improvements -- for the wow factor, UPS doesn't separate them out -- saved around 10 million gallons of gas and reduced emissions by the equivalent of taking 5,300 cars of the road for a year.
For fun I've used my GPS navigator for parts of the city I know well. I've not found its selected routes to be very optimal, though a couple times it suggested a better route I'd never thought of.
I suspect what it needs are 'hints' from human users familiar with the areas.
We could use data from other drivers' recent trips to suss out good routes. If I designed a GPS navigator I would add two buttons: Like and Dislike. Users would just tap one to indicate if a route seemed especially good or bad. I mostly want this for my cycling GPS but it could be useful for cars too.
It's already a GPS... why not just time the routes directly?
(Assuming you're concerned with time, which I assume is at least often the case. For convenience and other 'softer' amenities, hints make much more sense.)
At least in NYC, they don't ban left turns, but the majority of streets are one-way; and they alternate street by street (or avenue by avenue). So you often need to put together 3 rights to make the left. There are exceptions (e.g., 42nd, 34th, and 23rd are all two way streets); but not too many.
In my city, many largest streets have left turns prohibited at all minor crossings - if you're "getting off the main road" and need to be on the opposite side, then three rights is the only way to get there.
From the article: "While the no left turn rule has an appealingly simple and algorithmic quality to it, you will see UPS drivers take left turns on occasion, especially in residential neighborhoods without much incoming traffic."
Well, sure, when there's no traffic against which to have to turn, the cost of making a left is negligible -- especially in American residential neighborhoods, which are often laid out so as to prompt suspicion that they weren't intended to be navigable at all. (And this is not even entirely false! A city neighborhood with access from multiple traffic arteries, and with straight, minimal-distance routes between them, invites through traffic to employ those routes as shortcuts, which is tantamount to inviting rush hour into everyone's driveway.)
I'm just annoyed we have another article with an inaccurate title that doesn't actually have any examples in the article (like a map of different routes) that show the how and why. Its basically a UPS fluff piece with no substance.
It really bugs me because this something to show folks the value of math and analysis. What types of problem solving go on that can really be improved by application of maps and big data. Heck, even the linked article doesn't have a map explaining why this is important.
Not to defend the article too much but the title is accurate.
It's imprecise but accurate. "UPS routing algorithms have an extremely strong weighting against left turns outside of residential non-through streets" is well within the errors bars of "UPS trucks don't turn left" when using a tool like conversational English.
In fact the less data the article gives the more it should prefer "UPS trucks don't turn left" you shouldn't report falsely precise digits.
The article has the phrase I quoted that says the title is inaccurate. This is HN, shouldn't we at least have some meat and an article that doesn't directly say the title is wrong?
It wasn't false. "Why UPS Trucks Don't Turn Left" is not "Why UPS Trucks Never Turn Left." If a truck does not turn left at an intersection, you can ask why it didn't turn left, even if it has and will turn left at other intersections.
But really, who cares? What is the obsession with the titles?
When I initially saw the title, I was expecting it to be a mistake and actually say "Why UPS Trucks Don't Turn Right". I guess the problem is less severe in the rest of the world.