Untangling complex syste.., p.100

Untangling Complex Systems, page 100

 

Untangling Complex Systems
Select Voice:
Brian (uk)
Emma (uk)  
Amy (uk)
Eric (us)
Ivy (us)
Joey (us)
Salli (us)  
Justin (us)
Jennifer (us)  
Kimberly (us)  
Kendra (us)
Russell (au)
Nicole (au)



Larger Font   Reset Font Size   Smaller Font  

  lyzed inside a laboratory where it is easier to control variables and parameters. When we face Complex

  Systems, this is not always feasible. Whether the environment is the laboratory or the real world, it

  often plays a relevant role. In fact, the environment can influence the phenomenon under study (espe-

  cially if it obeys nonlinear laws), and it can also affect the performances of the observer, and the instru-

  ment. The use of an instrument is mandatory for the observer; the device allows making quantitative,

  objective, and reproducible determinations. Any measurement, however accurate and performed care-

  fully, is tainted by errors, and the results have inevitable uncertainties. There are two principal sources

  of errors and uncertainties: one is the instrument that we use, and the other is the operator.1 There are two types of sources of uncertainty: they are the systematic and the random errors.

  D.2 SYSTEMATIC ERRORS

  Any instrument determines the value of a variable with a definite degree of accuracy. When we

  choose an instrument to collect data, we must be aware of the error we make by using it. Almost

  all direct measurements require reading either a marked scale (e.g., the marked scale of a ruler, a

  chronometer, or a pipette) or a digital display (e.g., the digital display of a spectrophotometer).

  When we read a marked scale, the main source of error is the interpolation between the scale

  markings. For instance, let us measure the length of the clip shown in Figure D.2 with a ruler

  graduated in millimeters. We might reasonably decide that the clip’s length is undoubtedly closer

  to 3.2 cm than it is to 3.1 or 3.3 cm, but no more precise reading is possible. We conclude that our

  best estimate is 3.20 cm, and the uncertainty of our determination is in the range 3.15 and 3.25 cm.

  1 The contribution of the environment is included in those of the observer and instrument.

  517

  518

  Appendix D

  Environment

  Observer

  Instrument

  Object or

  phenomenon

  FIGURE D.1 Protagonists of a measurement.

  FIGURE D.2 Determination of the length of a clip by a marked ruler.

  The correct way to present the result of the measurement will be lbest = ( .

  3 20 ± 0. )

  05 cm. More in

  general, when we measure a quantity x , we collect our best estimate x and we specify the range

  true

  best

  (∆ x) within which we are confident x lies:

  true

  xbest ± x

  ∆ [D.1]

  The term ∆ x is the absolute error of our measurement, and it is usually stated with one significant

  figure. At most, with two significant figures, but no more. Moreover, the last significant figure in

  x must be of the same order of magnitude of the uncertainty, that is in the same decimal position.

  best

  If we want to increase the accuracy of our determination and reduce ∆ x , we need a finer instrument.

  For instance, in the case of the length of the clip, we may use a Vernier caliper, like that shown in

  Figure D.3. First, we close the jaws lightly on the clip. Then, we measure its length. In the Vernier caliper, there are two marked scales: a fixed one and a sliding one. The fixed scale is identical to the scale

  of the ruler of Figure D.2. The separation between two consecutive marks is equal to 1 mm. The sliding scale has 20 marks. The left-most tick mark on the sliding scale will let us read from the fixed scale

  the number of whole millimeters that the jaws are opened. In Figure D.3, it is between 3.2 and 3.3 cm.

  For the estimation of the clip’s length at the level of one-tenths of mm, we must determine the number

  of the first mark of the sliding scale that is aligned with a mark of the fixed scale. In Figure D.3, it is the seventh mark that corresponds to 0.30 mm. The uncertainty, as indicated onto the caliper, amounts

  to 0.05 mm. Therefore, the final best estimate of the clip’s length is lbest = ( .

  3 230 ± .

  0

  )

  005 cm. Note

  that the use of the caliper has increased the accuracy of our determination of one order of magnitude.

  The quality of a collected datum is expressed by the ratio of ∆ x over x , which is called the

  best

  relative uncertainty (ε rel):

  x

  ε rel = ∆ [D.2]

  xbest

  Appendix D

  519

  FIGURE D.3 Determination of the length of a clip by a Vernier caliper.

  FIGURE D.4 Reading of a scale whose marks are rather far apart. The distance between two consecutive

  marks is (1.85 ± 0.05) cm.

  If we multiply ε rel by 100, we obtain the percentage relative error. The percentage relative error

  is ε rel = 1 5

  . 6% with the ruler, whereas it is ε rel = 0 155

  .

  % with the caliper. We confirm that the use of

  the caliper reduces the uncertainty of one order of magnitude.

  Usually, the uncertainty in reading a scale when the marks are fairly close together is half the

  spacing. If the scale markings are rather far apart, like those of the pipette shown in Figure D.4, we might reasonably assume that we can read to one-fifth of a division. The volume of the liquid taken

  with the pipette of Figure D.4 is reasonably estimated to be (0.24 ± 0.02) mL.

  The error we make when we read a value on a digital display is usually of one unit on the last

  significant figure, unless indicated differently by the specifications of the instrument we are using.

  The errors described so far generated by the inherent inaccuracy of the instrument we use, are

  said systematic because they are always present, and their absolute extent (∆ )

  x is constant and

  known, even before making the measurement.

  An instrument may be a source of another type of systematic error that cannot be known unless

  we measure standard quantities, or we repeat our measurement by using different devices. Possible

  miscalibration of our equipment generates such a second type of systematic instrumental error.

  520

  Appendix D

  A

  B

  FIGURE D.5 Parallax error in reading a pipette’s scale. Case A represents the correct viewing angle,

  whereas case B represents the incorrect one.

  Moreover, the operator may introduce systematic errors in the measurements, as well. For

  instance, the right way of reading a volume on a scale of a pipette is to put our eyes at the height of

  the liquid’s meniscus, like in situation A of Figure D.5. If we read the volume systematically by viewing the meniscus from above (see situation B of Figure D.5), we determine the volume constantly

  by defect. In situation B of Figure D.5, we will read (1.40 ± 0.05) mL rather than (1.50 ± 0.05) mL.

  The introduction of such uncertainty, called parallax error, might be even worse depending on the

  angle of viewing and the type of scale.

  D.3 RANDOM UNCERTAINTIES

  It may also occur that sometimes we read a scale assuming the right position, but in some others

  not. In this case, not all the measurements are affected by the same parallax error, but only some of

  them. The uncertainty introduced in such a way is not systematic but random.

  The instrument itself may become a source of random errors. For instance, a chronometer may

  run faster or slower accidentally; a photomultiplier may change abruptly and temporarily its sensi-

  tivity to photons. Often, it is the environment that can contribute to the occurrence of random errors.

  In fact, fluctuations in the surrounding conditions may influence the performances of the instrument

  and the observer.

  The only way for estimating the contribution of the random error is to repeat the same measure-

  ment many times. For example, imagine determining the lifetime of a long-lived phosphor by using

  a chronometer. A source of error is our reaction time in starting and stopping our clock: if we delay

  in starting or we anticipate in stopping, we underestimate the lifetime; if we delay in stopping or

  we anticipate in starting, we overestimate its lifetime. Since all these possibilities are equally likely,

  Appendix D

  521

  our measurements will spread statistically. We will have a set of results: τ , , …, whose limiting

  1 τ 2

  τn

  distribution is Gaussian. The best estimate of τ true is the center of the Gaussian distribution (read

  Box 1 of this Appendix), which is the average when n → ∞:

  n τ i

  ∑

  τ

  i 1

  best = τ =

  =

  [D.3]

  n

  The uncertainty on the average2 is the standard deviation of the mean:

  1

  n τ( i −τ 2)

  n 1∑ i=

  σ

  1

  τ =

  −

  [D.4]

  n

  BOX 1 THE GAUSSIAN DISTRIBUTION

  The form of the Gaussian function describing how the measurements of the τ value are

  true

  (τ τ

  −

  2

  true)

  distributed over the variable τ is f (τ ) 1/σ π

  2

  e

  σ

  = (

  ) − 22 .

  Why is the average the best estimate of τ ?

  true

  If we collect n values ( τ , , …, ), the probability of getting is proportional to

  1 τ 2

  τn

  τ 1

  (τ

  2

  1 τ

  − true)

  1

  −

  f (τ ) ∝

  e

  σ

  2 2

  1

  σ

  π

  2

  The probability of obtaining τ is proportional to

  2

  (τ

  2

  2 τ

  − true)

  1

  −

  f (τ ) ∝

  e

  σ

  2 2

  2

  , and so on.

  σ

  π

  2

  (τ

  2

  n τ

  − true)

  The probability of getting τ is proportional to f (τ

  σ

  1

  2 ) −

  2

  2

  n ) ∝ ( /σ

  π e

  .

  n

  The probability of observing the entire set of n values ( τ , , …, ) is just the product of the

  1 τ 2

  τn

  probabilities of observing each of them when all the n measurements are independent events.

  n

  n

  (τ τ

  ∑

  −

  i

  i

  1

  1

   − =

  ttrue)2

  prob(τ

  2

  2σ

  1,τ 2 ,

  ,

  … τ n) ∝ f (τ1) f (τ2)… f (τ n) = 

  e

  

  

   σ

  π

  2 

  According to the principle of the Maximum Likelihood, the best estimate of τ is that value

  true

  for which prob(τ

  …

  τ

  …

  1 τ 2

  τ

  1,τ 2,

  ,τ n) is maximized. Obviously,

  (

  prob , , , n) is maximum if the

  exponent n

  ∑ (τ

  2

  2

  2 is minimum. If we differentiate the exponent with respect to

  =1

  i −τ true ) / σ

  i

  τ and we set the derivative equal to zero, we achieve that τ

  n

  best = τ = ∑

  τ n

  / .

  true

  =1 i

  i

  ( Continued)

  2 The uncertainty on the average value τ is the standard deviation of the mean, whereas the uncertainty on the single 2

  determination τ is the standard deviation σ =

  1

  n (

  ) .

  n 1 ∑

  τ i −τ

  i

  −

  i=1

  522

  Appendix D

  BOX 1 (Continued) THE GAUSSIAN DISTRIBUTION

  By using the principle of the Maximum Likelihood, it is possible to also find the best esti-

  mate for σ, which is the width of the limiting distribution. If we differentiate prob(τ

  …

  1,τ 2,

  ,τ n)

  with respect to σ, and we set the derivative equal to zero, we confirm that σ is the standard

  best

  deviation of the mean. In the figure within this box, there are two Gaussian distributions, both

  centered at the same τ , but having two different widths, i.e., two different

  true

  σ values.

  0.9

  ( τ − τtrue)2

  −

  f ( τ) = 1

  2 σ 2

  e

  σ√2 π

  0.6

  σ = 1

  f ( τ)

  0.3

  σ = 3

  0.0 5

  10

  15

  τ

  d.4 ToTal uncerTainTy in direcT measuremenTs

  In synthesis, all the direct measurements are subjected to both random and systematic errors (see

  Figure D.6). Some random errors are generated by the observer, whereas some others by the equip-

  ment. The unique strategy to cope with them is to repeat and repeat, as much as possible, the

  experimental determination. Also, the systematic errors may derive from both the observer and the

  instrument. The remedy consists in changing the operator or the experimental device. Whatever is

  the accuracy of a device, we cannot completely wipe out the systematic instrumental error due to

  the inherent uncertainty of any apparatus.

  This means that our measurements of x can never be represented as dots along the x-axis

  true

  (see Figure D.7) but as circles, whose area shrinks by upgrading our experimental apparatus.

  Figure D.7 shows four possible situations that may arise when we determine through direct mea-

  surements the value x of a variable. Case (1) occurs when the random error (

  ) is small and

  true

  εrandom

  when the only type of systematic error ( ε ) is that generated by the inherent uncertainty of the

  sys

  instrument. In fact, the measurements are represented by small circles distributed symmetrically

  with respect to x and pretty close to it. Such determinations are said to be accurate and precise.

  true

  Case (2) comes about when the random error is small, the inherent uncertainty is tiny, but our

  measurements are affected by miscalibration. In case (2) the determinations are still precise ( ε

  random

  is small like in case [1]) but they are inaccurate (the overall ε is large). Case (3) refers to a situa-

  sys

  tion where our measurements suffer from scarce precision, although they are accurate because the

  instrument has small inherent inaccuracy and there are no problems with miscalibrations. Case (4)

  represents the worse situation among all because the distribution and size of the circles along x-axis

  reveal that our determinations are imprecise (they are spread in a broad range of x values) and inac-

  curate (they are big circles, all located at values smaller than x ).

  true

  Appendix D

  523

  Change

  Change

  instrument

  Miscalibration

  operator

  Observer’s

  systematic

  errors

  Inherent

  instrumental

  inaccuracy

  PHYSICAL

  QUANTITY

  Operator’s

  Instrumental

  random

  random

  errors

  Repeat

  Repeat

  errors

  FIGURE D.6 Types of uncertainties in a direct measurement and their remedies.

  (1)

  xtrue

  (2)

  xtrue

  (3)

  xtrue

  (4)

  xtrue

  FIGURE D.7 Four possible situations we may encounter when we measure x more than once.

  true

  When we know both the random ( ε

  ) and the systematic uncertainty (

  ), and they are com-

  random

  εsys

  parable, the reasonable estimation of the total uncertainty will be the sum of the two contributions:

  xbest + x

  ∆ = x + ε random + ε sys = x +σ x + ε sys [D.5]

  From equations [D.5] and [D.4], it is evident that if we repeat many times our measurement, i.e.,

  then n → ∞, ε

  can become very tiny and negligible with respect to

  . However, even if we

  random

  εsys

  repeat an infinite number of times our measurement, the total uncertainty in x cannot be smaller

  best

  than ε that is the inherent inaccuracy of our experimental device.

  sys

  d.5 ProPagaTion of The uncerTainTies

  Many variables are measured indirectly. They need to be calculated from the values of variables

  that have been measured directly. For instance, the speed of a vehicle requires the direct determina-

 

Add Fast Bookmark
Load Fast Bookmark
Turn Navi On
Turn Navi On
Turn Navi On
Scroll Up
Turn Navi On
Scroll
Turn Navi On
183