Embedded Files
Uncertainty is everywhere
Uncertainty is everywhere
Engineers need to deal with uncertainty. There is no such things an infinite precision. We still need to make stuff.
Need methods and tactics to cope with and design knowing that uncertainty exists
Get used to always considering uncertainty. Make it like a reflex. If you are quoting a numerical value that relates to something in the real world Stop! and think if it needs an uncertainty quoted at the same time. If you are a manager, reject findings that don't indicate the confidence the author has in the results. It might be that they have lots, or little, but need to be established and, at best, quantified.
There is stuff from uncertainty lab video here.
Some key definitions
Some key definitions
The terms ‘error’ and ‘uncertainty’ are not the same, although they are often confused.
The terms ‘error’ and ‘uncertainty’ are not the same, although they are often confused.
As we can never make a ‘perfect’ measurement, all measurements can only be estimates of the 'true value’. Error is the difference between the measured value and the ‘true value’ of the thing being measured.
Since there is always a margin of doubt about any measurement, uncertainty is how you can communicate this uncertainty. It is a quantification of doubt.
Errors are normally classified in three categories: systematic errors, random errors, and blunders.
Errors are normally classified in three categories: systematic errors, random errors, and blunders.
A systematic error is a consistent error, where the same influence affects the result of each repeated measurement.
They are usually caused by faulty equipment or a flawed experiment design.
If you repeat the experiment, you’ll get the same error.
A random error is caused by unknown and unpredictable changes in the experiment. These changes may occur in the measuring instruments or in the environmental conditions.
They are unpredictable and can’t be replicated by repeating the experiment again.
A blunder is an outright mistake. You may record a wrong value, misread a scale or forget a digit.
If you take multiple measurements, a blunder should stick out like a sore thumb.
Can you ever know the true value?
Can you ever know the true value?
It is impossible to measure any physical quantity with complete certainty. From random variations in the quantity being measured to limitations in the measuring equipment or the experimental method, there will always be errors in any measurement.
This means that if we measure some quantity and then repeat the measurement, we will almost certainly measure a different value the second time.
Any measurement result must be accompanied by a statement of the uncertainty in the measurement. Only when this uncertainty is stated, can the fitness for purpose of the measurement be properly judged.
Report abuse