How Accurate Are Luggage Scales At Airports?

[Renato1]

If expect a professional scale would be cleaned, tared to zero, and then checked with a known (say 20.0kg) weight regularly.

So would I, especially when penalties may be involved.
Not doing so is like the Police issuing speeding tickets using uncalibrated cameras.

But as my example at Doha airport demonstrates, either it isn't being done, or the frequency of doing it isn't adequate.
Regards,
Renato

 

[Vic]

What isn't known is whether the +/- 0.3 or 0.5kg relates to putting a standard weight on the scales in the same spot, or to people putting their feet in different positions on the scale - which does have an effect on accuracy, as you rightly point out.

All I know is that I have three bathroom scales at home, one of which consistently reads 1.6kg less than the highest reading one, and the third reads in between the other two. Thus no way of having confidence as to which is giving the right answer.

Yet by measuring myself, with and without luggage, and being very careful not to move the scales, and to put my feet on exactly the same position, I get repeatable results off each machine matching each other - which vary within a range of around 0.3kg. And more often than not, that weight figure accords very closely with the scales at the airport - which is why I get suspicious at airport check-ins the less frequent times when the weight figure from my technique is significantly different from theirs.

I used the difference method with bathroom scales in italy two and a half weeks ago. I measured the parcel we were about to send as being 9.4 kg. The scales at the post office had it weighing 9.2kg.

I think that's pretty accurate for most purposes.
Regards,
Renato

P.S. - I forgot to mention one thing in relation to the accuracy of bathroom scales. While they do often have tolerances of +/- 0.3kg or +/- 0.5kg written on their boxes or sometimes on a label on the unit, if one repeatedly steps on them in the same foot position and weighs one's self, the weight results only usually vary by 0.1kg on some units to 0.2kg on other ones. It's not like the results vary by 0.6kg or 1kg, as the tolerances would suggest.

This is the difference between precision and accuracy. Precision being the ability to measure the correct value and accuracy the range of variation in the measured value. It is possible to very accurately measure a 20 kg, +/- 0.001 kg, as 18 kg. an Accurate measurement that is not precise.

With a bathroom scale showing an accuracy of +/- 0.3 or 0.5kg, by measuring twice the accuracy of your final number is automatically +/- 0.6 or 1kg. The quoted number must also only be about the mechanical accuracy of the scale. Additional uncertainty is introduced by other external things like positioning on the scale. In terms of holding luggage in changes the weight/mass distribution. Unless you're holding in centered above your head.

 

[Renato1]

This is the difference between precision and accuracy. Precision being the ability to measure the correct value and accuracy the range of variation in the measured value. It is possible to very accurately measure a 20 kg, +/- 0.001 kg, as 18 kg. an Accurate measurement that is not precise.

With a bathroom scale showing an accuracy of +/- 0.3 or 0.5kg, by measuring twice the accuracy of your final number is automatically +/- 0.6 or 1kg. The quoted number must also only be about the mechanical accuracy of the scale. Additional uncertainty is introduced by other external things like positioning on the scale. In terms of holding luggage in changes the weight/mass distribution. Unless you're holding in centered above your head.

Let's suppose I have a set of bathroom scales with stated tolerance of +/- 0.5kg, and I have a suitcase I think is 20kg.

I stand on scales myself and I weigh at 90kg (accuracy +/- 0,5kg).
I stand on scales myself and suticase and I and it weigh at 110.2 kg (accuracy +/- 1 kg)
I stand on scales myself again and I weigh at 90.1 kg (accuracy +/- 1,5kg).
I stand on scales myself and suitcase again and I and it weigh at 110 kg (accuracy +/- 2kg).
I stand on scales myself again and I weigh at 89.9kg (accuracy +/- 2,5kg)
I stand on scales myself and suitcase again and I and it weigh at 110.1 kg (accuracy +/- 3kg).

I keep getting close to repeating results, yet by your analysis my uncertainty is greater than ever, so much so as to render the results unreliable.

The tolerance is the same at each step, it doesn't increase with each weighing. And the scales' measurements repeat better than the stated tolerance. With my digital bathroom scales with tolerance of +/- 0.5kg, it means that if I weigh myself and it says 90kg, the actual weight may be say 90.1kg and it will consistently give results around that reading by a lesser amount than the stated tolerance.

Though the old spring analog bathroom scales I used to have did vary on repeat by half a kilogram or more. Luckily, that's not the case now,
Regards,
Renato

 

[Vic]

Let's suppose I have a set of bathroom scales with stated tolerance of +/- 0.5kg, and I have a suitcase I think is 20kg.

I stand on scales myself and I weigh at 90kg (accuracy +/- 0,5kg).
I stand on scales myself and suticase and I and it weigh at 110.2 kg (accuracy +/- 1 kg)
I stand on scales myself again and I weigh at 90.1 kg (accuracy +/- 1,5kg).
I stand on scales myself and suitcase again and I and it weigh at 110 kg (accuracy +/- 2kg).
I stand on scales myself again and I weigh at 89.9kg (accuracy +/- 2,5kg)
I stand on scales myself and suitcase again and I and it weigh at 110.1 kg (accuracy +/- 3kg).

I keep getting close to repeating results, yet by your analysis my uncertainty is greater than ever, so much so as to render the results unreliable.

The tolerance is the same at each step, it doesn't increase with each weighing. And the scales' measurements repeat better than the stated tolerance.
Regards,
Renato
Renato

It doesn't work like that, each of the above measurements have an uncertainty of +/- 0.5 kg.

If you then take your first measurement pair and subtract, the suitcase is 20.2 kg +/- 1 kg.
Same error applies to each individual set of measurements = +/- 1 kg

If you want to get a bit tricky and use the average of the three measurements then the error becomes Sqroot(3) x 0.5 = 0.87 (3 is the number of measurements)
So your average is 90 +/- 0.87 kg and 110.1 +/- 0.87 kg

Then the suitcase is 20.1 +/- 1.7

Making a single measurement of the suitcase alone just has the error associated with the device used to make that measurement.

 

[lovetravellingoz]

All I know is that I have three bathroom scales at home, one of which consistently reads 1.6kg less than the highest reading one, and the third reads in between the other two. Thus no way of having confidence as to which is giving the right answer.

ANSWER: All will be inaccurate. Especially with bathroom scales, and moreso if they have never been re-calibrated. But all weighing devices will be out by some amount.

Also the variation will probably vary subject to the mass weighed. One may be more accurate with say 20kg, and less accurate if weighing a person holding bag. The other one or two may be vice versa.

In addition the tolerance will most likely not a be a uniform +- amount over 20-120kg. The +- will probably be greater at the higher weight and so weighing yourself and the bag together increases the likely error. Perosonally to cross check I normally put the bag directly on the scales or just on top of a small board.



If I am flying economy, and especially on budget airlines, I will aim to be under and not on any mass limit.

 

[Renato1]

It doesn't work like that, each of the above measurements have an uncertainty of +/- 0.5 kg.

If you then take your first measurement pair and subtract, the suitcase is 20.2 kg +/- 1 kg.
Same error applies to each individual set of measurements = +/- 1 kg

If you want to get a bit tricky and use the average of the three measurements then the error becomes Sqroot(3) x 0.5 = 0.87 (3 is the number of measurements)
So your average is 90 +/- 0.87 kg and 110.1 +/- 0.87 kg

Then the suitcase is 20.1 +/- 1.7

Making a single measurement of the suitcase alone just has the error associated with the device used to make that measurement.

Suppose I go to a National Measuring laboratory and find out my weight is 90.2kg. If I then weigh myself on bathroom scales and the scales say I am 90kgs, the tolerance means the true weight is somewhere between 89.5 to 90.5kgs - consistent with the accurate measurement.

Not being a precision instrument I repeatedly weigh myself on the same scales, and consistently get readings of between 89.9 and 90.1, averaging 90kg. I then hold a 30kg suitcase and the scales say 120kg. True weight is somewhere between 119.5 and 120.5kg. I repeat the weighing, and measurements vary by between 119.9 and 120.1, averaging 120kg.

I find the difference between the two averages to be 30kg for the suitcase, and estimate the error as at +/- 0.1kg. It doesn't matter that I am actually 90.2kg.

And this is a lot better than the the estimated error of +/- 0.5kg from a single reading of the luggage on the scales.
Regards,
Renato

 

[Renato1]

ANSWER: All will be inaccurate. Especially with bathroom scales, and moreso if they have never been re-calibrated. But all weighing devices will be out by some amount.

Also the variation will probably vary subject to the mass weighed. One may be more accurate with say 20kg, and less accurate if weighing a person holding bag. The other one or two may be vice versa.

In addition the tolerance will most likely not a be a uniform +- amount over 20-120kg. The +- will probably be greater at the higher weight and so weighing yourself and the bag together increases the likely error. Perosonally to cross check I normally put the bag directly on the scales or just on top of a small board.



If I am flying economy, and especially on budget airlines, I will aim to be under and not on any mass limit.

The question isn't whether the scales are accurate, but whether the difference between their measurements with and without luggage are accurate. Tolerance will likely be uniform between 90kg and 120kg, which is the range I measure at.
Regards,
Renatoi

 

[Vic]

Suppose I go to a National Measuring laboratory and find out my weight is 90.2kg. If I then weigh myself on bathroom scales and the scales say I am 90kgs, the tolerance means the true weight is somewhere between 89.5 to 90.5kgs - consistent with the accurate measurement.

Not being a precision instrument I repeatedly weigh myself on the same scales, and consistently get readings of between 89.9 and 90.1, averaging 90kg. I then hold a 30kg suitcase and the scales say 120kg. True weight is somewhere between 119.5 and 120.5kg. I repeat the weighing, and measurements vary by between 119.9 and 120.1, averaging 120kg.

I find the difference between the two averages to be 30kg for the suitcase, and estimate the error as at +/- 0.1kg. It doesn't matter that I am actually 90.2kg.

And this is a lot better than the the estimated error of +/- 0.5kg from a single reading of the luggage on the scales.
Regards,
Renato

Just outlining the scientific approach to propagation of uncertainty in measurements. The error in your averages will be the square root of (the number of measurements) multiplied by 0.5. You can just decide the error is 0.1 kg

If you get weighted by the national measurement laboratory they will tell you you're 90.2 kg +/- X kg. X likely being small. There is no point then getting a different measurement for yourself on the bathroom scales. That's just a different new measurement with it's own uncertainty. Just go weight yourself and suitcase on the bathroom scales, now the uncertainty is +/- (0.5 + X) kg.

You might also be missing the point about precision. If the scale consistently shows 130 kg for something that is actually 125 kg, then if can be very accurate, but not precise. Accuracy is about consistency of measurement, precision is about the ability to get the correct value. Some airport scales that vary from your measurement might be very accurate, but also not precise. It's hard to say if you're measurement is correct or the airport measurement is correct. But both can be accurate

And that's it from me...

 

[zenyattamondatta]

If you want to get a bit tricky and use the average of the three measurements then the error becomes Sqroot(3) x 0.5 = 0.87 (3 is the number of measurements)
So your average is 90 +/- 0.87 kg and 110.1 +/- 0.87 kg

You should be dividing by the square root to get the error in the estimated mean, not multiplying.

 

[Vic]

You should be dividing by the square root to get the error in the estimated mean, not multiplying.

Yes, that's what you do to calculate the standard error of the mean, divide the standard deviation of the mean by the square root of N. But in this case +/- 0.5 kg is not the standard deviation of the mean it is the uncertainty in each measurement.

 

[zenyattamondatta]

So you're saying averages will have more error in them the more measurements you take.

I'm sorry, but you're just wrong here.

 

[Renato1]

You might also be missing the point about precision. If the scale consistently shows 130 kg for something that is actually 125 kg, then if can be very accurate, but not precise.

That is exactly what I have been saying in every post.

If the scales consistently measure me at 95kg when i actually weigh 90kg, and the scales consistently measures me and the luggage at 125kg when we actually weigh 120kg, then the difference between the two readings is very accurate - more accurate than the stated tolerance on the scales.
Regards,
enato

 

[Beano]

I work in an industry which needs calibtated scales. I weigh my carry-on before I leave work on profesionally calibrated scales, especially if travelling with Jetstar. I've rarely been questioned on Qantas when travelling international but often in domestic, which has never been a problem, 14 kg (7 +7, two bags). This time flew Jetstar and was weighed after security and found to be over??? I had to go back and check in one bag ( I had 20KG check-in). Once upon a time did the same but bags came in well under wieght at check-in so all OK. Checked them with another weighing machine in the same area and found they were well over! The Jetstar sclaes were well off. In preactice found check-in scales reasonably accurrate, even against digital or my old dial scales. Bathroom scales will vary significantly in accuracy and repeatability. Newer digital types may be OK, but old analogue not so.

 

[Beano]

PS. Repeatability does not necessarily mean accuracy unless calibrated.

 

[shambles]

If anyone’s interested enough in the question “How often are check-in luggage scales calibrated?”, my better-half works in the same FedGov department that holds all the standards for measures & does stuff like checking petrol-bowser accuracy. National Measurement Institute.

Just to add to Forg's insider perspective, I also work in the weighing industry and as per above, certainly all "official" airport scales in any modern economy, are supposed to be NMI (or equiv Gov. body) approved as Legal For Trade scales because a financial transaction is potentially levied based on the scale's judgement. This is exact same as petrol pumps, parking meters, retail scales etc. The same rules do not necessarily extend to "courtesy scales" which you can find scattered around airports in boarding gates etc. However, QANTAS over the last year or so have been replacing the coughpy old dial scale hand luggage frames with more up to date digital versions (made by the Japanese company I work for) and they decided to make sure these new scales are also NMI approved which means they are considerably more expensive than non-NMI approved scales. There was an interesting article in the SYD paper a while back, about exactly this topic......
Read the article here
If your personal scales differ to these official scales I'd bet my left and right ones that the airport scales are correct!

Personal Scales vs Proper ones - who is right?
One thing that you often hear though (including that article), is punters talking about their bathroom scales or hand held luggage scales. Frankly speaking, those type of scales are cheap & cheerful technology usually made in a Chinese toy factory. You are talking sub $100 items vs NMI approved scales that would typically sell for closer to a grand. We also supply the often controversial boarding gate scales that JQ use. These are commonly the source of angry discussions from peeps who insist that their scales at home know better. Bottom line with all these official scales where excess fees can be charged, is that a licensed weighing company is engaged to check and certify them at least once a year. (For JQ & QF that's us) These are industrial grade scales and the passive environs of an airport are heaven for them compared to the harsh industrial or freight depots where they typically get used. The only way any of these would be weighing incorrectly would be if they sustained serious damage/shock-loading (vehicle or similar). If they were found to be damaged, they'd be swapped out and repaired which also requires re-verification.

Check-In counter built in Scales
The built in scales you see at the check in counters are mostly made by a globally successful NZ company called ATRAX. They specialise in airport scales and they are absolutely all NMI approved and you'll see them all over the world. The scales at self check-in counters are also often made by these guys. As per above, these scales are checked & re-calibrated at least every 12 months and you should see a calibration sticker with a date and the local licensed company that's done it. The masses used for checking and certifying scales are also certified and traceable.

Global Standards
Pretty much all countries that hail from the "British empire" would follow exact same rules and process regarding legal-for-trade weighing instruments. In Australia it's now the role of NMI not just to approve an instrument for trade use (that approval process can typically cost us as scale manufacturers about $25K per model so it's not cheap!), but they are also responsible for policing Weights & Measures compliance. 10 years or so ago, the latter was responsibility of state Gov Weights & Measures officers but like a lot of things, this was reduced and centralised so in reality, not much policing happens any more. There are NMI bodies all over the world and most now have mutual recognition agreements. Certainly Japan, Aus, Europe & UK all recognise and approve products for/with each other. The US runs a similar system called "NTEP".

Hope that helps inform the "masses" or at least cure insomnia

 

[TonyD]

Hope that helps inform the "masses" or at least cure insomnia
[/QUOTE]

I dont doubt any of that, nevertheless JQ portable scales at OOL were 2.4 kg or 15% different to check in scales at JQ OOL and JQ SYD . or the personal operating them was cheating ?

 

[AustSerbKor]

I've had check-in scales under weigh my bags flying PUS-KUL-OOL on D7, I placed the big bag on the scale and it read 7-8kg for each of 5 large suitcases.
Unfortunately, I pre-paid 120kg so I didn't win on that one and I'm sure it wouldn't have happened if I hadn't have prepaid.