91Èȱ¬

My comment is not very economicsy, but is based on queues. It concerns stop/go signs at roadworks - you know, where you have some roadworks blocking one of two opposing lines of traffic, and traffic going one way or the other is directed past the obstruction, usually in turns. I heard that the "most efficient" method of setting the stop/go sign was to never alter the direction of traffic allowed through. This avoids the inefficiency of having to wait for one line of traffic to finish going past the roadwork before the opposing line gets going. It also has the implication that half the traffic will never go anywhere! Is it true? Does it have any NHS analogies? (eg fix broken right legs, but not broken left legs.)

Of course, if you only have access to one ticket operator, you have to try something else to reduce the queue - like running the trains at different times, or adding fast ticket machines.

Evan, are you suggesting that the NHS be privatized? Good job you're not a politician, eh?

Both pricing and queuing are forms of rationing, and both have costs (it costs to employ a person for the extra hour in the morning, to have two working ticket booths, etc). Which form of rationing is preferable depends on balancing overall costs/benefits, the alternatives available (rail queues v. road queues) and equity beliefs on who should bear costs and benefits.

That balance will vary between industries and customers, and it is a bit one sided to imply the economic argument is as clear cut as your articles have implied.

A major reason for the development of queues is under-pricing of the services sought. Thus we have devices such as higher peak/lower off-peak transport prices so that those with less need to travel at peak time have an incentive not to. The problem is most acute with "free" (i.e., fully tax-funded services), where the only cost potential users need to consider is that of their time. In Australia, there has been rapid growth in use of private education and health services in recent years, with users forgoing the services they have paid taxes for in favour of buying non-government services with perceived higher quality and/or lower delays. The incentives for good service are higher for private providers than public ones.

Crucially however, the length of queue affects the amount of people joining in virtually all situations (the NHS included, presumably).

If the queue for operations was shorter, more people would join. The queue is an equilibrium mechanism much like a price whereby people trade off the cost of queueing with the benefit of what they get at the end of the queue.

I agree and it can be applied to anything really. Even down to the queue in our office for the printer in the morning. It would make sense to either invest in a second printer or get a faster one in the first place!

I agree wholeheartedly that the majority of cases where regular queues develop could be solved at very little expense. Hiring a ticket officer for just 2 hours each morning would cost maybe £20 - a very small ticket price price spread over all the passengers - but save loads of people's queuing time.

I wonder if there is a cultural dimension to this too: are people in the UK by and large happier to wait for services than those in the USA? If so then rational service providers will spend money to shorten their US operations queues rather than their UK ones.

There is one problem with your station in the real world, where do you get the one person to work one hour one hour and what do they do for the rest of the time. This may not seem very hard to arange. This is possible if it is always the same time every day and last just the one hour but in the real world this hardly happens and any time over the one hour means they have less time doing their own job and that can get demoralising. It can get even harder when you have training and holiday cover to concider. Back to the NHS I don't undersand why they don't do operations twentyfour hours a day this at least bring down equipment cost.

I wish the managers of Alliance and Leicester could have thought about this when they remodelled their Croydon branch with only two available spaces for cashiers. To serve the 330,000 population of Croydon.

Robin Linacre says that the longer the queue, the less people likely to join. This might be true for buying an newspaper in the shop or something

However not everyone has the choice- if you need a hip, you have to join the queue, there isn't much alternative. only way queue will get shorted if the poor lady dies.

Oh i suffered from a chest infection that lasted like 6 weeks because i didnt want to wait for the GP- in the end i had to go as i just wasn't getting better- so all i did was defer the wait. I still eventually added to the GP's queue but in addition, my economic output declined due to the infection time. ( of course , with the internet and web 2.0, i don't actually do much work, but thats a different matter)

There is one trick which seems to be used sometimes to 'shorten' treatment queues, and that is the introduction of 'assessment' queues which have to be negotiated before reaching the actual treatment queue. Of course the patient doesn't necessarily benefit from this, but it makes the actual measurement of waiting times somewhat subjective. I'm reminded of the many changes to the definition of 'unemployment' which were used by past governments to manipulate their figures.

The equilibrium is maintained in the way you describe as long as there's an affordable alternative. Most people waiting for a major operation on the NHS have no alternative but to join the queue. Hence the queue gets longet and longer.

You are quite right that this does not just apply to the NHS. The logic of the queue is also the reason that carbon-offsetting schemes do not work as a solution to carbon emissions.

Think of the carbon dioxide in our atmosphere as a "queue". Carbon offsetting, for example by planting trees, removes a fixed amount of carbon from the queue much more slowly than, say, the trans-Atlantic flight that put it there in just 8 hours. So while we all continue jetting about, thinking that we can buy our way out of our emissions problem, the queue actually continues to get longer.

Because of the mismatch in the rates of carbon joining and leaving the "queue" in the atmosphere, all offsetting does is very slightly reduce the rate of queue growth. But it makes a nice profit for those who administer such schemes and allows sanctimonious rock stars to delude themselves!

I think the point about the NHS is not that the more clinicians should be employed on a temporary basis to reduce the queues (although that would be nice).
The point is that by delaying operations the queue gets longer and the trusts save a bunch of money in that year. But the next year the queue equalibrium is restored, they don't save any money, but the queues remain long.

Short term benefit with long term problems.

The NHS has finite resources. However, unlike the private sector, regardless of how many patients it treats the financial pot is the same. The management of queues is based on flexibility. Movement of staff to meet demand. More difficult in areas of high specialisation. However, much can be done in dealing with those who abuse the system, for example, those who fail to attend appointments, those who demand treatment for minor ailments, those who abuse the staff. Punitive financial penalties for such abusers would significantly deter such people thus reducing queues and allowing time to be targeted on those in genuine need. Not to mention swelling the coffers of the NHS.

It would be interesting to extend the analysis here into a stochastic framework.

Evan Davies argues that queues stay constant when the flow of new arrivals equals the flow of people being served.

This is obviously and trivially true as stated.

However consider the problem of a (slightly more realistic) situation in which the arrival rate is, on average, the same as the departure rate, on average.

If the queue is currently length 10 then for any time in the future the expected length is 10. However the process defined has no mean reversion (i.e. once the queue length has increased to 11 the expected length at all points thereafter is 11, or once it has declined to 9 the expected length at all times thereafter is also 9.

As time goes on the probability of seeing a queue of any arbitrarily long length approaches unity.

Of course once we have done that we need to also start to add individual behaviour to the problem. The probability of someone choosing to join a queue is not independent of the length of the queue which does introduce mean reversion.

The author of this article is very lucky if the queues in the supermarkets he goes too are as short as he says. Whenever I go grocery shopping the queues are horrendous and have only got worse since the introduction of these stupid 'self-service' tills that aren't actually operated by a human.

The problem with the train ticket analogy which you have not mentioned is the fact that you may save *MY* time by having more employees selling tickets. The costs are bourne by the train operator and the benefits are enjoyed by me. If you are trying to make a profit then why bother?

Maybe we should have special ques with a guaranteed minimum waiting time in return we may 10% more for our tickets? In this way the economic interests of the train operator are aligned with the interestes of the consumer.

(A classic example of this is the 0870 customer service numbers, the longer they put you on hold, the more money they make and the worse the customer service is, customers interests are not aligned with service providers profit.)

queing is a british thing, most europeans are abit more pushy and firey, I think it is shame when my mum and mum in law who both worked for the nhs as nurses have had to resort to pay privately £25000 for a life threatening double bypass and £15000 for a hip replacement they had paid into the system all their lives.
Perhaps we should become even more
Americanised and have a private health scheme where everybody pays in to their own scheme and they all get seen when they need the treatment and not at the whim of the pct's

Of course there is the fact that on the NHS some of the people IN the queue and some JOINING the queue do not make it out alive.

Is your local station Cheltenham Spa by any chance, it sounds like it?

The problem with your argument is it assumes the 'queue controller' cares about the 'queuers' waste of time.

As regards to the rail station, if you had recognised mobile ticket sellers (on a commission) they would soon turn up at busy times and start selling tickets.

Of course, there is another way to prevent queueing and that is to persuade people to go away. A relative of mine on an NHS waiting list for an operation which would significantly reduce the pain she was in was asked by phone whether as she had been waiting so long they could now take her off the list. I don't think I will comment further, but you can imagine my indignation at the time and cynicism when NHS Trusts announce reduced waiting times.

In response to Robert Linacre, I would suggest that whether the number of people joining a queue depends on the queue length depends on whether people have a choice about using the service in question.

To expand upon Evan's example, if there is a long queue to buy a newspaper at the station, then consumers may choose not to join that queue, but the same cannot be said for tickets as they are required. I would consider that people seeking hospital treatment are almost exclusively there through need and not desire.

2. Evan, are you suggesting that the NHS be privatised?

This is probably the undertone, given that Evan is the author of a book called "Public Spending" (1998) in which he argues the case for the use of private companies in the supply of public services.

It's still floating around on Amazon: 

You owe me comission Evan!

About time we dumped this Stalinist fossil.
What other business would solve a cash flow problem by doing less work.

Whilst I cannot fault the logic, it ignores the fact that the rail queue may shorten by people giving up and going home, or seeing the queue and being put off joining it. In the case of the NHS, the sadder fact of people dying whilst waiting, or going private. Both acts of leaving the queue.

Hence the queue has additional costs above that indicated in the article and may not infact result in a permanently longer queue if people leave it.

Other than that, a great article and nice logic.

Erlangs! that brought back memories of my time in Telephone exchanges. One Erlang was one circuit in use for one hour. It was not economical to have enough circuits through the exchange to cope with the busy hour ( 10-11am in my Strowger exchange days ).We usualy provided less so that most of the circuits where not idle for most of the day.

NHS queues have an additional complicating factor. As the queue gets longer people start to drop out before they get their operation because:
1) They get better without treatment
2) They get fed up of queing and pay to go privately.
3) They die either of the disease that they are waiting to be treated for or of old age.

This oversimplified analysis completely ignores the key point.

A system with zero queues is the most inefficient system you can have from the point of view of the provider. If the NHS had no waiting lists, and you just had to rock up at the hospital to have an operation, that would mean that every department in every hospital had a surgeon just waiting around for someone to turn up. But then, while you were being operated on, then someone else might turn up needing the operation. So a queue forms. So you need more doctors waiting around just in case that happens. And in fact, just in case there is an epidemic of a particular problem, you need a huge surplus of doctors (and beds, equipment, space, etc.) in every department, in every hospital.

This is an extreme case of course. In reality the ideal is that every operating theatre has a short queue (but it is essential that there is a queue). And this is a very tough balance to get right because conditions are fluctuating all the time - the number of people needing a given operation at a given time is very unpredictable.

In terms of pure costs (this is a very cold inhuman calculation!), the surplus doctors cost a fortune, while the queues cost nothing except bad publicity and a bit of admin.

What's the point of queue theory, in this respect, without a theory of employment markets and why people work in the first place?

One of the major problems of the NHS, for example, is matching the demand of the customer and what people are willing to work/what the employer is prepared to do to make them work.

The only solution for the NHS, as it currently works, is to get people to work in a way they currently refuse to do. And that's not a trade union problem; it's a problem that supermarkets can overcome because they employ people who are willing work at almost anytime and have few bargaining abilities or rights.

It's all there in the final comment. There has to be a direct incentive to those responsible for managing queue length in order for them to do so.

And the inherent inefficiencies of the public sector means lacks the accountability/reponsibility to do so. And this wastefulness doesn't just apply to queues.

Queues are a consequence of monopoly power. There's only one train, doctor, dentist, and so on.
Monopolists take no account of the external costs they create amongst their customers. In a competitive market, suppliers don't cause customers to spend their time/money on waiting, or they'll lose income.
Doctors are scarce: that's why they don't like shortening the training period. They want to retain that scarcity.
Moreover, doctors have asymetric information in their favour: they own the diagnoses and the patients' medical records.
Paying hospitals & GPs only for work done, and to a fixed price, is one way of breaking this monopoly.

one solution to the queue problem is to issue tickets - so that when your ticket comes up you can get your service done.
it won't make the queue any quicker but you won't have to stand uncomfortable close to someone else infront and behind you.

the english tend to spontaneously form queues for no good reason (like 5 people queuing for a bus that they know will be totally empty - but still queueing)

maybe they like it

Is the assumption that anyone that reads this is completely stupid and needs patronising and spoon-fed about simple principles?! I got sick of reading this blog way before the end: 'queues get longer if more people join it than leave it'.. smarter than the average bear!
I'd become accustomed to Mr. Davis' voiceover on Dragons' Den pre-empting or repeating exactly what each Dragon says but had thought that was down to the scriptwriters assuming the people watching are imbeciles.. maybe I was wrong!

Is the assumption that anyone that reads this is completely stupid and needs patronising and spoon-fed about simple principles?! I got sick of reading this blog way before the end: 'queues get longer if more people join it than leave it'.. smarter than the average bear!
I'd become accustomed to Mr. Davis' voiceover on Dragons' Den pre-empting or repeating exactly what each Dragon says but had thought that was down to the scriptwriters assuming the people watching are imbeciles.. maybe I was wrong!

It's a nice argument. Unlike supermarket queues, where the worst that can happen is customers become inconvenienced, with the NHS people die while in in the queue. Is this considered a lost "customer" to the NHS? Perhaps more market forces, akin to supermarkets, is what the NHS needs.

Evan,

I don’t think it’s nitpicking to point out that the ‘rules’ about queues are not really rules; they’re just properties of queues once they’re seen as things that can change in length. But that apart, it is true that, when a queue starts to lengthen, increasing appropriately the rate at which queuers’ needs are met can stabilise the length of the queue. However, the apparent huge excess of benefits over costs in your hypothetical example rests on the assumption that the queue would eventually stop lengthening without any action – if the rate of joining will eventually fall to equal the rate of leaving anyway, then a once-and-for-all measure introduced while the former exceeded the latter is then sufficient to eliminate the queue. But has this any applicability to the NHS?

NHS managers charged with balancing the finances can’t be expected to shorten queues by taking into account personal benefits accruing to patients through shorter waiting lists, since the benefits – if they could be measured (as opposed to guessed at) at all – don’t show up in their books. That’s surely the responsibility of government when they set NHS budgets. But is the heart of the problem not the way in which the principle of a health service ‘free at the point of delivery’ is implemented in the UK? At the moment it involves a funding mechanism that conceals from the patient the cost of individual treatment and thus the benefit of a healthy lifestyle – so there is no incentive to be careful. It entertains the prospect of being able to meet any demand placed on it – that, were it actually taken seriously, would require a level of taxation that would probably be politically unacceptable. The healthy would have reasonable grounds for opposition as their resources are presently taken away through general taxation to meet the needs of the unhealthy – seemingly justified by the notion that the likelihood of being ill is equal for all, irrespective of lifestyle. In this context are queues not more efficient than expecting NHS managers to perform miracles without money? The healthy might say it is also fair!

Here's an example of why Evanomics (and even Economics) doesn't apply to french administration.

To pay for my daughter's school dinners, I should go to the town hall once a month to pay for the month ahead. Usually, the queue is atleast 40 minutes long.

In november, I decided to stop wasting my time, and haven't been to pay since.

They still haven't asked me to pay a penny.

Since november 2005.

And they're still feeding my kids.

NHS queues are very different to other types of queues as there is more than one way to get out of the queue. You can:

1. Get to the front
2. Die
3. The condition can spontaneously resolve
4. The condition can worsen beyond the proposed treatment
5. You can get fed up and go private

The longer the queue, the fewer people ultimately end up with option 1 which costs the NHS. And the more people end up with 2-5 which save the NHS (apart from perhaps 4). So longer queues can save NHS money in the short and long term

I'm not sure how one could apply this to the NHS, perhaps someone could come up with an answer and is based partially on my anecdotal evidence.

Sometimes planning only seems to cause greater queues. For example, traveling through a busy jucntion at rush hour. The lights and flows are normally planned but despite this queues occur. However, when there is a power failure and the lights are off people seem to sort themeslves out and the queues do not appear as long.

Applying this to the supermarket example we see a little of both at work. On the one hand an adequate number of check-outs need to be open but once so we sort ourselves out by looking for the shortest queue.

I seem to remember that this theory has been researched before but I cannot think of how (if at all) it could be applied to this case, particularly given the NHS is there to serve the sick and weak.

There is a major difference between supermarkets and the NHS which I believe is critical.

The supermarkets are a service provider who can gain a benefit from having their customers NOT queue, in the form of loyalty, repeat visits and increased sales. This gives them the incentive to invest the costs of providing the additional service and keep queues at an efficient minimum.

The NHS are a service provider for whom there is no such incentive. They are not concerned with efficiency, simply with spending. This is because the health 'market' is inefficient and not perfectly competitive.

Even when the service rate exceeds the arrival rate, queues still form because arrivals aren't evenly spaced - there will be times when many people arrive and have to queue and times when few people arrive and the server is idle.

In a simple one-server queue where arrivals and departures after service happen at random, keeping the number of people in the system (being served or waiting) at 4 or less on average means the server has to be idle at least 20% of the time.

Since this amount of idle time is hard to justify to service providers, many of them have reduced service levels and thus longer queues on average.

Reducing queue length is only cost-effective for the servive provider if they attach a cost to queue size. Perhaps this could be opportunity loss if the queue is very long, or loss of queuers taking their custom elsewhere.

The problem with queues is simply the problem with the idea of a "rush hour" if everyone tries to do the same thing at the same time then you will need maximum capacity for say two hours, and then no capacity at all! This kind of capacity boosting would require us to momentarily turn half of the country into a hospital, or a road.
Queuing allows us to spread out the bumps and spread capacity over time instead of space. That is not to say however that the rush hour idea is always bad: The bittorrent download system works on the idea that each person brings part of the capactity with them, as is the idea with self-service checkouts, but these limitations are still there, in different forms.
When you are dealing with many different needs, efficiency and uniformity do not go together.

Using the railway station analogy, the other method of reducing the queue is to promote all those people paying cash for daily travel passes to the front of the queue as they can be served quicker then people requiring annual season tickets, which require a photo-card and a credit card transaction.
I understand that this method has already been employed by the NHS, treating people for tonsilectomies or wart removal to reduce overall waiting lists, putting those requiring more complicated procedures - such as hip replacements - to the back of the queue indefinitely.

At the end of your post you mention that supermarkets cut queues to remain competitive, in some instances businesses are happy for certain customer groups to queue.

Take airlines for example. Airlines are quite happy for economy passengers to spend a long time in a queue to check-in and board the aircraft, so they can differentiate their high margin business class products by offering dedicated check-in desks and priority boarding lanes.

If you make people wait longer they may get better on their own or they may die.Alternatively they may get fed up and decide just to live with their ailment. So the extra waiting time may not increase the waiting list too much.
However if somebody is on a waiting list they may be off work so it makes sense to treat them quickly so they can get back to work.
It's not a simply a question of normal queuing. There are personal apsects to consider.

So, what's the difference between supermarkets and the NHS??

Supermarkets are in a very competitive market where the price differentials are small and customers can easily switch. So they have a strong incentive to keep queues short and shelves stocked.

The NHS has such a huge price advantage over it's competitors that the vast majority of customers can't afford to take their business elsewhere.

No-one has mentioned the concept of fast-track queues for simple transactions. Supermarkets often have "baskets only" queues which allow you to be processed more quickly if your transaction will be short. It's a shame that banks don't do this: many times when I go to pay in cash and cheques, I get stuck behind people who want to open an account without the right documentation (and with a poor command of English) or who have a complicated problem to solve. Having one cashier dedicated to simple transactions would work wonders to reduce the queues. Sadly many service organisations don't treat queues as important: often my bank has staff standing around behind the scenes chatting or in meetings, rather than having a simple rule that reducing the length of queues takes precedence over everything else.

Supermarkets are a poor analogy. A cashier is a relatively unskilled person (at least I was when I was a cashier many moons ago) and any cashier can serve any customer, which means that meeting increased demand is purely an issue of staff volume.

The issue with the NHS is that meeting demand requires staff with very particular skills. Hiring more GPs across the board does not help the queue for brain surgery. Add to this the fact that skilled labour is being drawn out of the marketplace by foreign and private clinics, and you have a problem.

Evan, your "3 rules of queuing" is in fact just one rule with two variables. And I thought you were an economist!

I used your theory about 10 years ago, to save our company money. When a customer queried an invoice of ours, it took us 2 weeks, before we looked at it (because of a backlog). We reduced this to 1 day, simply by a one-off effort. Since then, query turnround has remained at 1 day, so we've had 10 years worth of getting queried invoices paid 9 days earlier. That makes the cost of the overtime, 10 years ago, seem very good value. The sad thing is, it took 2 years to get it past our directors, because they didn't understand the theory. Seeing the same theory, with a "91Èȱ¬ stamp" on it, gives it more authority, and I'm sure many companies could utilise it.

Your comment has a point but one flaw.
IIf at the train station there is only one ticket machine that issues the tickets than employing one more person would not solve the problem. The train company would have to employ an extra person AND buy an extra ticket machine.
So what happens to the extra machine when the queue is at such a state that it doesnt require a 2nd person? It sits there being unused.
Ticket machines may not be that expensive and the purchase of one may be easily justified. However, X-Ray machines, CAT Scanners and the like are a tad more expensive, not only including the purchase price, but the upkeep and whatever storage costs are asociated with them and therefore justification for purchase of these additional pieces of kit would be more difficult to come up with.

I think many people are missing Evan's point.
As described in the Sunday Telegraph link, hospitals "are now deliberately delaying operations to save money"...."despite staff and equipment being available".
So right now, we have lots of doctors and nurses stood around doing nothing, while the queues get longer. It might well save the NHS some money this year, but its a ridiculously short-sighted strategy!!

#53: Evan is an "economist" not a mathematician. I wonder if Evan even reached a conclusion in his rant?

There are some points to make:
Is it possible to have a competitive health system? Privitisation is a factor but - in the case of the USA - is even more inefficient than state-subsidised, in terms of financial cost as a percentage of GDP or real cost per person.

Waiting lists on the NHS have many contributory factors. If someone needs a new heart, then someone else has to die to make that heart available. Perhaps this constraint will be overcome by using stem cells and somehow having heart farms in the future. In the case of an heart transplant queue, there are actually many queues (a queue for each heart type) but all going through the same check-out (i.e. the surgeons and operating theatre).

A combination of the latest equipment and theatres, with well-trained staff, plus a quick turnaround time (think Ryannair) would help the NHS. Perhaps future Goverments and Health Authorities will transfer more of the budget towards this, rather than towards middling-management and unnecessary operations? Perhaps future medical advances will speed things up and reduce post-op complications and hospitalisation?

1. Shorter queues do not necessarily mean a shorter wait for the persons queued merely because the time taken to serve the person in front may well exceed that of the person served at the longer queue in front of the next window/cash point, as is the case in supermarkets with older people needing help with their money, people asking for additional information or shoppers over-buying with credit cards refusing payment.
2. One wonders what it cost to put in/build all those service windows at the station, supermarket or NHS if no one is likely to use them. Would the money invested be more or less than hiring extra staff, even part-time?
3. In Italy queuing up for ages to be served is tradition.
The other day at the post office one employee walked back into the office leaving his window to make a very loud and lengthy personal phone call despite 15 of us queued out in front.
The lady in charge when I complained excused him by saying he’d had an car accident and needed to speak to his insurance company! Just how many of us can afford to walk off the job like that?
Had I insisted I could be reasonably certain that my incoming and outgoing mail would suffer the consequences.

To make matters worse, our small post office has introduced a numbering system meant to cut waiting time BUT with 2 out of 4 windows open only for specific operations (like posting letters/parcels) despite it being possible to make payments there, even if the staff are twiddling their thumbs they insist on serving customers as per the randomly selected numbers on the display. To complicate matters there are 4 different types of slips issued on the ticketing machine, so either you take one of each to be sure you can get the first free window, or you wait and wait and wait... If you’ve taken the wrong number back you go!

4. As for the National Health service here everyone seems to have a friend or relation working there who can miraculously and completely eliminate the need to queue. The few who don't like me can wait and suffer or die. The latter often make the 8 o'clock news.

I think it somewhat disingenious to compare waiting times at a train station and the NHS. Then state that 'to solve a one year deficit crisis you might be permanently stuck with longer waiting times.' The two systems are clearly in different phases; the ticket office is a stationary system with no new resource input to allow for more customers to be served and therefore queues to be shortened. The NHS is current going through an increase in investment under which waiting times are falling and have not yet reached a new equilibrium. Any short term reverse in waiting times due to end-of-year financial targets should only be temporary until a new equilibrium is reached. Only then is it likely that any delays may become permanent.

Operating theatres are expensive to build, outfit and maintain and it is right that they should be used as much as possible. But under the current system employing more staff to use them overnight will have an increased cost which will have to be cut from some other service the NHS provides. If people want their access to NHS medical care to improve there are only two methods to do this. 1) Provide more resources (ie cash) either by increased taxes or by personal charges to those taking up treatment, or 2) increase the efficiency of the use of resources already paid for.

The NHS is limited in what it can do by how much money it is given by the population of this country in any one year. Unlike the railways and the supermarkets there is no mechanism for the NHS to recoup the extra cost of doing more work. The supermarket or train operators can simply raise their prices, the NHS cannot. Also, contrary to some previous posts, the NHS is not a monopoly. Those who wish to avoid the queues can bear the full cost of their treatment by going private. It is not until you examine the costs of private treatment that you realise just how much your trip to the GP is actually worth in hard cash.

Not the point I know:

The answer to the train and any other queue is ensure all of us have a "black box" implanted at birth and then we can be billed for whatever we do and, of course, "managed" properly.

Queuing theory is well understood within the three classic queuing models Erlang A,B and C and an understanding of these models means you will recognise the phenomena that many of the other contributers refer to. Such as resource utilisation. The Erlang A model is probably the closest to hospital waiting lists.

From the queuing models you can deduce that there are a number of ways of reducing queuing in the NHS.

1) Reduce service time. Can the logistics of the operation be improved. Are the operations carried out as fast as possible (without reducing safety or quality) perhaps there is delay in getting the equipment to the right place or a wait for the right operating theatre.

2) Reduce the demand for the operation.
Can the problem be avoided by preventative care or treatment.
say dietary advice or earlier intervention that reduces the need for the operation in the population.

3) Pool the queues. Several hospitals in an area should share the same queue for an operation. Or thedemand should be fed by a single super hospital. This will increase the resource utilisation - and reduce the size of queue needed to have a certain resource utilisation % increasing efficiency. Although it will increase the distance patients may have to travel. Of course closing hospitals in order to concentrate resources is usually not popular.

4) Increase the number of servers.
Kind of obvious that the more resource available to do the operation then the shorter the queues will be. Queue length is VERY sensitive to the number of servers.
The resource provision level should seek to balance the cost of extra resource again the gain of reducing the queuing time. The correct level of resource should mean we have short BUT NOT ZERO length queues.

One thing no one considered was the ability of one person to do multiple jobs... thus Multitasking! Not sure if this would go well witht he NHS but any other case this might work well. Especialy the Train Ticket offices. We could simple destribute the resources as needed and one sector of company starts to get little above reasonable drafting someone with capable abilites from anoter area to compensate for a short time... Now witht he NHS we could have maybe paramedics doing some of the nurses chorse. (example) I am not aware of the full services functions but this could be plausable...

To be honest, I don't have a massive understanding of the way in which queuing works within the NHS. However, I don't really see it as a queue in any meaningful way that can be compared to a supermarket, a bank, or a train station. The 'queue' to me is more like a puddle where you have no understanding of how many people are in front of you, how long they might take to be dealt with, or even when you're next. Add to that the fact that people are then entering the 'queue' perhaps ahead of you because they are deemed more urgent, or as someone else said people die, but again you have no visibility of this in the same way as you do in the supermarket. Is it really a queue anyway???

Regardless, this stuff about saving money by not performing procedures despite having access to the appropriate resources seems like madness to my untrained eye. Talk about false economy. 'Let's let everyone get sicker!' Right, because long term, that won't cost any more will it??? Duh!

I read way back up there a little post about the Alliance and Leicester in Croydon which made me chuckle as I've heard this complaint so many times. Someone back there mentioned how cost and queuing times affect the way people queue - well this is exactly what the A&L are doing. They are (very) reluctant to start charging you to do your banking over the counter (remember ATM charging), so instead they make your like as difficult as possible by making you queue for an alarming length of time. They also make their employees lives miserable - but hey - they don't like it, they can join the unemployed queue. It's not just A&L, it's the Halifax, HSBC and probably loads more. They want you to use the internet because it's cheaper for them which equals more profit.

Banks are not service providers - I'm sure they once dressed it up that way, but I'm not sure it was ever true. (Or it ever will be......)

I studied queuing theories at University - fascinating stuff!

Supermarkets can afford to pay minimum wage staff to sit around if their queue dwindles to zero, surgeons/wards/operating theatres are a bit more expensive to lie dormant.

Individual queues look short (and hence attractive), but when you are next in line and they start paying by cheque, counting pennies out one-by-one etc. you can get fustrated. That is why many banks/post offices make everyone queue and you get the next available counter.

"...the queues cost nothing except bad publicity and a bit of admin."

They often cost weeks/months/years of suffering for the patient, along with a lot of time off work (a cost to the country) and, in the worst cases, they cost lives.

#55: "Your comment has a point but one flaw.
IIf at the train station there is only one ticket machine that issues the tickets than employing one more person would not solve the problem. The train company would have to employ an extra person AND buy an extra ticket machine.
So what happens to the extra machine when the queue is at such a state that it doesnt require a 2nd person? It sits there being unused.
Ticket machines may not be that expensive and the purchase of one may be easily justified. However, X-Ray machines, CAT Scanners and the like are a tad more expensive, not only including the purchase price, but the upkeep and whatever storage costs are asociated with them and therefore justification for purchase of these additional pieces of kit would be more difficult to come up with."

This would be a good point if there was ever a time when the NHS was in a state of readiness to treat someone. We're pretty far from being in a situation when doctors and nurses have nothing to do but sit around and wait for patients. (Although, unfortunately, they too often do seem to sit around and do nothing)

The current parlous financial state of the NHS, combined with services that have shown little improvement despite all the extra billions has proven without any doubt that the NHS is in it's death throws and Labour's chronic mishandling has hastened its demise.
As Even points out, a business would not survive with the queues seen in the NHS.
When will people wake up to the fact that the NHS is not "free", it costs us all a lot of money that would be better invested in a different system, such as a state backed insurnace system with a choice of service provider

The rail network suffers from loss of revenue through ticketless travel, some of which will no doubt be time-pressed commuters abandoning their queue place when their train arrives at the station.

Sadly, a similar routine doesn't seem to be in place for patients on NHS waiting lists :-)

Evan's article and every one of the subsequent 67 posts has overlooked one significant factor.

Making a comparison between queueing for a train ticket and waiting for an operation does not take account of the how people use the waiting time.

In the ticket queue you'll twiddle your thumbs and stare into space, which makes it dead (and expensive) time to you, but waiting for an operation is not dead time. It's not as if there is absolutely nothing you can do while waiting: even if you're signed off work, you can still make good use of the time (educate yourself using written or audio coursework, see more of your family/friends etc.)

The issue is much less simple than it first appears.

No, to solve the queue problem, we must provide sufficient service providers and create an environment, whereby paying customers are willing to give up more for a quicker, improved service. Its simple. Making the service that is so heavily used, (either because it is preferred or easily accessible) more scarse will improve the queue.