A method to consider a maximum admissible risk in decision-making procedures based on measurement results ACTA IMEKO ISSN: 2221-870X June 2023, Volume 12, Number 2, 1 - 9 ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 1 A method to consider a maximum admissible risk in decision- making procedures based on measurement results Alessandro Ferrero1, Harsha Vardhana Jetti1, Sina Ronaghi2, Simona Salicone1 1 Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB), Politecnico di Milano, Italy 2 Dipartimento di Energia (DENG), Politecnico di Milano, Italy Section: RESEARCH PAPER Keywords: measurement uncertainty; threshold; decision making; risk of wrong decision; maximum admissible risk Citation: Alessandro Ferrero, Harsha Vardhana Jetti, Sina Ronaghi, Simona Salicone , A method to consider a maximum permissible risk in decision-making procedures based on measurement results, Acta IMEKO, vol. 12, no. 2, article 41, June 2023, identifier: IMEKO-ACTA-12 (2023)-02-41 Section Editor: Laura Fabbiano, Politecnico di Bari, Italy Received March 17, 2023; In final form May 31, 2023; Published June 2023 Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Corresponding author: Simona Salicone, e-mail: simona.salicone@polimi.it 1. INTRODUCTION Measurement results are very often used as input elements in decision-making procedures, which represent the core element of conformity assessment. This is a very critical task in many fields, from the industrial one, where conformity of a product's feature to given specifications must be assessed, to environment protection, health, legal and forensic ones, where decisions are generally related to checking that the presence of a substance (a pollutant, a drug, etc.) or the error of an instrument does not exceed a given threshold or maximum admissible limit. Most decisions – if not all of them – are taken by comparing a measurement result with a threshold or a range of admissible values, where the threshold, or the upper and lower limits of the range, are given as simple quantity values [1]. Then, according to where the measurement result is located with respect to the threshold or the range, a decision is taken on whether conformity can be declared or not. If measurement uncertainty is not considered, or if it can be assumed to be negligible, this decision can be easily taken by comparing two numerical values: the measured value with the threshold (as shown in Figure 1) or the measured value with the upper and lower limits of the range. Figure 1 shows that, in such a situation, the decision is apparently taken with no risk of being wrong. However, even if measurement uncertainty has been evaluated and found to be negligible, a risk of wrong decision still exists, because it is widely recognized [2] that “when all of the known or suspected components of error have been evaluated and the appropriate corrections have been applied, there still remains an uncertainty about the correctness of the stated result, that is, a doubt about how well the result of the measurement represents the value of the quantity being measured”. It is also well-known, according to the GUM [2], that in many applications “it is often necessary to provide an interval about the measurement result that may be expected to encompass a large fraction of the distribution of values that could reasonably be attributed to the quantity subject to measurement. Thus, the ideal method for evaluating and expressing uncertainty in measurement should be capable of readily providing such an interval, in particular, one with a coverage probability or level of confidence that corresponds in a realistic way with that required”. When measurement uncertainty is taken into account, again a decision about conformity can be readily taken if the coverage interval is completely below or above the threshold (Figure 2). ABSTRACT Measurement uncertainty plays a very important role ensuring validity of decision-making procedures, since it is the main source of incorrect decisions in conformity assessment. The guidelines given by the actual Standards allow one to take a decision of conformity or non-conformity, according to the given limit and measurement uncertainty associated to the measured value. Due to measurement uncertainty, a risk of a wrong decision is always present, and the Standards also give indications on how to evaluate this ri sk, although they mostly refer to a normal probability density function to represent the distribution of values that can be reasonably attributed to the measurand. Since such a function is not always the one that best represents this distribution of values, this paper considers some of the most-often used probability density functions and derives simple formulas to set the acceptance (or rejection) limits in such a way that a pre-defined maximum admissible risk is not exceeded. mailto:simona.salicone@polimi.it ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 2 On the other hand, the situation represented in Figure 3 appears to be quite critical, since the threshold falls inside the coverage interval representing the fraction of the distribution of values that could reasonably be attributed to the quantity subject to measurement (the measurand). This means that there is a probability that some of the values that could reasonably be attributed to the measurand might be greater than threshold t, even if the measured value x̅𝑚 is lower than the threshold. This also means that, if conformity shall be assessed when the measurand is lower than the threshold, a risk exists of declaring the measurand conforming while it is not, and this risk can be evaluated starting from measurement uncertainty [3]. Conformity assessment involves, therefore, a decision- making process affected by uncertainty. Such a problem has been widely covered in the literature [4]-[6], mostly by taking epistemic uncertainty into account [7]. However when the input elements to a decision-making process are measurement results, uncertainty takes a well-defined meaning, defined by the VIM [1] and the GUM [2], and such a definition and the related evaluation methods cannot be disregarded when evaluating the risk of wrong conformity assessment, as clearly shown in [8]-[13]. This problem is covered by the BIPM document JCGM 106:2012 [14], in a very extensive way under a strict metrological perspective, and treating uncertainty according to the GUM recommendations [2]. In particular, it covers the problem of stating whether a measured quantity falls inside a given tolerance interval, which is defined in [14] as the “interval of permissible values of a property”. According to the above definition, the tolerance interval can be both a closed interval and a one-sided interval. Furthermore, document [14] defines acceptance limits in such a way that, given a measurement uncertainty value, the measurand is declared conforming if the measured value falls inside the acceptance limits and non-conforming when it falls outside these limits. The document considers different decision rules and the way to evaluate the associated risk of incorrect assessment starting from measurement uncertainty. Hence, it represents a very useful guide in evaluating the probability of declaring as conforming an item that is not, and vice versa. Although this problem is well discussed in [14] from a theoretical perspective, little guidance is provided, from a more practical point of view, on how to set the numerical value of the acceptance limit not to exceed the maximum admissible risk of making a wrong decision (once the measurement uncertainty and the maximum admissible risk are given). This is an important issue when dealing with critical measurements, such as those performed to protect health and environment. This paper, after having quickly reviewed the most used decision rules, proposes a method that, given a threshold (or more in general a tolerance limit), provides the acceptance limit as a function of measurement uncertainty and a predefined maximum admissible risk of exceeding the given threshold. Example are given for some of the most used probability distributions. 2. THE MOST COMMON DECISION RULES To correctly evaluate the risk associated with decision rules, it is necessary to identify or assume the probability density function (PDF) representing the distribution of values that could reasonably be attributed to the measurand [2], since this risk can be evaluated only after integrating such PDF from - to the threshold [2], [3]. It is well known that, according to the GUM [2], the standard uncertainty u(x) associated with a measurement result x is the standard deviation of the PDF representing the distribution of values that could reasonably be attributed to the measurand. On the other hand, the expanded uncertainty U(x) = ku(x) identifies a coverage interval [x – U(x); x + U(x)], built about the numerical value x of the measurement result, whose coverage probability depends on the assumed probability density function and the considered coverage factor k. It is also worth reminding that the PDF representing the distribution of values that could reasonably be attributed to the measurand depends on the available information. It is generally – and wrongly – considered that the available information comes only from the employed measuring equipment [15], while document JCGM 106:2012 [14] states that such information always has two components: the one available before performing the measurement (called prior information) and the additional information supplied by the measurement. The resulting, or posterior PDF, can be obtained by applying Bayes' theorem [14]. Keeping in mind the above considerations, it is possible to consider and discuss the two most common and employed decision rules in conformity assessment. It is assumed that the PDFs considered in the following sections are always the posterior PDFs. 2.1. Decision rule based on simple acceptance This rule, also known as shared risk, considers accepting as conforming (and reject otherwise) an item whose property has a Figure 1. Comparison of a measured value with a threshold, when the measured value 𝑥𝑚 is lower (a) and greater (b) than threshold t and measurement uncertainty is not taken into account. Figure 2. Comparison of an uncertainty interval with a threshold, when the uncertainty interval is completely below (a) and above (b) threshold t. Figure 3. Comparison of an uncertainty interval with a threshold, when the threshold falls within the interval. ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 3 measured value inside the tolerance interval. In this case, uncertainty is not explicitly considered. In mathematical terms, and assuming that the tolerance interval is given by all values lower or equal than threshold t, an item is accepted as conforming if the measured value xm of property x satisfies to condition 𝑥m ≤ 𝑡. Let’s make a few considerations about this decision rule. It can be readily checked that, assuming a symmetrical PDF about xm for the values that could reasonably be attributed to the measurand, the highest probability of exceeding the threshold is obtained in the limit case of xm = t and is 50%, no matter on the evaluated uncertainty value and the PDF. Therefore, when this decision rule is applied, measurement uncertainty does not affect the risk: reducing uncertainty only decreases the width of the interval of non-conforming values 𝑥nc that are considered as conforming, but does not reduce the risk of misidentifying non- conforming items as conforming, which still remains 50% (when xm = t). To define a maximum width of the interval of non- conforming values that are considered as conforming, a mutually agreed maximum acceptable expanded uncertainty Umax is generally set and it is therefore suggested that the expanded uncertainty U associated to the measured value, for a coverage factor k = 2, must satisfy U  Umax [14]. 2.2. Decision rule based on guarded acceptance / rejection The simple acceptance rule reported in Sec 2.1 shows that the closer the measured value to the threshold, the higher is the probability (up to 50%) of accepting an item as conforming that is not, and vice versa [14]. This probability can be reduced by setting an acceptance limit inside the tolerance interval, as suggested by [14] and as shown in Figure 4, when, respectively, the measured value is required to be lower or equal a given threshold (𝑥m ≤ 𝑇𝑈) - as in Figure 4a - and the measured value is required to be within a closed interval (𝑇𝐿 ≤ 𝑥m ≤ 𝑇𝑈) - as in Figure 4b. Figure 4 represents the case of guarded acceptance [14], that is the decision rule for which the risk of accepting a non- conforming item is reduced by setting an acceptance limit AU inside the tolerance interval (see Sec. 8.3.2 in [14]). According to this rule [14], if the tolerance interval is a one- sided interval, upper limited by TU (Figure 4a), an acceptance limit AU is set inside the tolerance interval. The interval between AU and TU (highlighted in yellow in Figure 4a) is called the guard band and its width (with sign) is defined as [14]: 𝑤 = 𝑇U − 𝐴𝑈 . (1) In the case of Figure 4a, it is w > 0. On the other hand, in the case of a two-sided tolerance interval, two acceptance limits AL and AU are set, as shown in Figure 4b. In this case, two guard bands are obtained, whose widths are defined as wU= TU – AU> 0 and wL= TL – AL< 0 respectively. From Figure 4, it can be conclusded that, when a guarded acceptance decision rule is considered, an acceptance interval smaller than the tolerance interval is obtained. This decision rule is hence in favour of increasing the probability that an accepted item is truly conforming. For the sake of completeness, let’s consider that, with respect to the two cases shown in Figure 4 (𝑥m ≤ 𝑇𝑈 and 𝑇𝐿 ≤ 𝑥m ≤ 𝑇𝑈), other two cases exist, that is: • 𝑥m ≥ 𝑇𝐿: in this case, considering guarded acceptance, AL is on the right of TL and w= TL – AL< 0; • 𝑥m ≤ 𝑇𝐿 ∪ 𝑥m ≥ 𝑇𝑈: in this case, considering guarded acceptance, AU is on the right of TU and AL is on the left of TL, so that wU < 0 and wL> 0; so that the obtained acceptance interval is smaller than the tolerance interval. A similar, though opposite situation is obtained in the case of guarded rejection [14]. In fact, this decision rule is in favour of increasing the probability that a rejected item is truly non- conforming [14]. In this case, acceptance limits are set, providing acceptance intervals greater than the tolerance interval. Without entering the details, as an example, by considering again Figure 4a, if the guarded rejection decision rule were applied, then the acceptance limit would be at the right of TU, thus providing a wider acceptance interval. In general, |w| is set as a multiple of the expanded uncertainty: |w|= rU [14]. If the PDF representing the distribution of values that could reasonably be attributed to the measurand is known or assumed, it is also possible to evaluate the risk of declaring a non-conforming value as conforming (or vice versa), as shown in Figure 5 in the case 𝑥m ≤ 𝑇𝑈, when a normal PDF is considered and |w| = U = 2u is taken, as suggested by [16]. In particular, Figure 5a and Figure 5b represent, respectively, the decisions of guarded acceptance and guarded rejection. Figure 4. Decision rule based on guarded acceptance. In Figure 4a a one-sided tolerance interval upper limited by TU is considered, while in Figure 4b a two- sided tolerance interval is considered between a lower and an upper limit TL and TU. Figure 5. Example when 𝑥m ≤ 𝑇𝑈 and a normal PDF is supposed. The standard uncertainty 𝑢(𝑥) and the maximum admissible limit TU (red line) are given in Table 1, and |𝑤| = 𝑈(𝑥) = 2 𝑢(𝑥) is supposed. The coloured area represents the probability of exceeding TU, when the measured value corresponds to AU, in the cases of guarded acceptance (a) and guarded rejection (b). (b) Tolerance interval w > 0 Acceptance interval Acceptance interval Tolerance interval wL < 0 wU> 0 (a) Guard band ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 4 To understand the relationship between w and the risk of wrong decision, let us consider the numerical example in Figure 5, where the maximum admissible limit (MAL, or, employing the same notation as the one in [14], TU) for a pollutant in water is assumed to be 50 mg/l. The pollutant concentration is assumed to be measured with a standard uncertainty of 5 mg/l, and the PDF representing the distribution of values that could reasonably be attributed to the measurand is assumed to be normal, as summarized in Table 1. According to these values, the expanded uncertainty with k = 2 is U = 10 mg/l. A guard band w = U = 10 mg/l is considered, so that an acceptance limit AU = TU – w = 40 mg/l is set, when guarded acceptance is considered. Therefore, the concentration of the considered pollutant in water is considered as conforming for every measured value xm  AU. If xm = AU, the situation shown in Figure 5a is obtained, in which the red line is located on TU. Since a coverage factor k = 2 has been considered, the coverage probability of interval [xm – U; xm + U] is p = 95.45%. Therefore, the risk of exceeding TU, that is the probability pw of taking the wrong decision, when this decision rule is adopted, is 𝑝𝑤 = 1−𝑝 2 = 2.28 %, independently of U. Of course, if xm < AU, pw < 2.28 %. On the other hand, if the aim of the measurement procedure is to assess, with high probability, that the pollutant concentration is higher than TU, the acceptance limit should be set, according to [14], at AU = TU + w. With the same numerical values and assumptions as before, this means that AU = 60 mg/l. Therefore, the concentration of the considered pollutant in water is considered as non-conforming for every measured value xm  AU. If xm = AU, the situation shown in Figure 5b is obtained, in which the red line is again located on TU. In this case, the risk of declaring that the pollutant exceeds the tolerance limit TU while it does not is again pw = 2.28%. On the other hand, the risk that the pollutant is above the limit is, obviously, 97.7%, 3. THE RELATIONSHIP AMONG UNCERTAINTY, ACCEPTANCE LIMIT AND THE MAXIMUM ADMISSIBLE RISK The example shown in the previous section relates uncertainty, acceptance limit and the risk of wrong conformity assessment in an implicit way, since it assumes that the distribution of the values that could reasonably be attributed to the measurand obeys to a normal PDF and the generally used coverage factor k = 2 is considered. These assumptions lead to the well-known 2.28 % risk of wrong decision. However, different situations with different PDFs and different values for the acceptance limits may occur in practical cases, where also different values might be desired for the maximum admissible risk (MAR) of wrong decisions. Therefore, a general formulation relating uncertainty, acceptance limit and MAR would be very useful to obtain one of 1 It is worth reminding that a normal PDF is generally obtained when the combined standard uncertainty is obtained as a combination of a sufficiently high number of contributions, so that the Central Limit Theorem applies, as suggested by the GUM [2]. On the other hand, the triangular and them, given the other two ones. Document JCGM 106:2012 [14] provides some very general indications on how to do this, mostly referring to a normal PDF. Attempts were made in the past, especially in the legal metrology domain [17], to set the acceptance limits in such a way that, given the measurement uncertainty, a pre-defined risk could be granted. However, to the Authors' knowledge, no practical indications are available to relate acceptance limits, measurement uncertainty, and risk in such a way that, having set two of them, the third one could be found. Such a relationship can be obtained starting from the PDF p(x) of the distribution of values that could reasonably be attributed to measurand x. Having defined such a PDF, the pertaining cumulative probability distribution function (CDF) can be obtained as: 𝐹𝑋(𝑥) = ∫ 𝑝(𝑡) d𝑡 𝑥 −∞ (2) It can be readily checked that 𝐹𝑋(𝑥) represents the probability that variable X is lower than x. Similarly, 1 − 𝐹𝑋(𝑥) represents the probability that variable X is greater than x. Therefore, using the same notation as the one used in [14], in the general case shown in Figure 4b, given a CDF 𝐹𝑋(𝑥), a tolerance limit TUL, and a MAR, if, for the considered measurable property, the measured value must be below the given tolerance limit TU, the following inequality must be satisfied: 𝐹𝑋(𝑇U) ≥ 1 − MAR , (3) while, if the measured value must be above the given tolerance limit TL, the following inequality must be satisfied: 𝐹𝑋(𝑇L) ≤ MAR . (4) Therefore, one of the following two equations must be solved to get the value of the acceptance limit AU (or AL) that ensures that the probability that the tolerance limit TU (or TL) is exceeded is exactly equal to the MAR: 𝐴U|𝐹𝑋(𝑇U) = 1 − MAR , (5) when the measured value must be below the threshold, or 𝐴L|𝐹𝑋(𝑇L) = MAR , (6) when the measured value must be above the threshold. The acceptance limit values ensure that, respectively, if 𝑥𝑚 ≤ 𝐴U (𝑥𝑚 ≥ 𝐴L), then pw  MAR, where xm is the measured value and pw is the probability of exceeding the tolerance limit, that is, the probability of wrong decision. Of course, solving these equations is strictly related to the shape of the PDF associated with the measurement result, and a solution cannot always be found in closed form. This does not prevent, however, the application of this method, because a numerical solution can be obtained by means of a Monte Carlo simulation, following the recommendations provided by Supplement 1 to the GUM [18]. On the other hand, the vast majority of the practical cases consider normal, uniform, triangular or trapezoidal PDFs1. In such cases, a closed-form solution can be readily obtained for AU (or AL), and, hence, the normal, uniform, triangular and trapezoidal PDFs are considered in the following. trapezoidal PDFs are generally obtained when two uniform PDFs are linearly combined, as in many practical measurement applications. Table 1. Numerical example. Max. admissible limit TU Standard uncertainty u pdf type 50 mg/l 5 mg/l Normal ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 5 3.1. The measurement results distribute according to a normal posterior PDF When a normal PDF is considered: 𝑝(𝑥) = 1 √2 π σ2 e −(𝑥−𝜇)2 2 σ2 , (7) where 𝜇 is the mean value and 𝜎 is the standard deviation. Then, the corresponding CDF is given by: 𝐹𝑋(𝑥) = 1 2 [1 + erf( 𝑥 − μ √2 ⋅ σ )] , (8) where: erf(𝑧) = 2 π ∑ (−1)𝑛 ⋅ 𝑧2𝑛+1 𝑛! ⋅ (2𝑛 + 1) ∞ 𝑛=0 (9) is the error function and can be well approximated with no more than 10 terms in (9). 𝐹𝑋(𝑇U) can, therefore, be written as: 𝐹𝑋(𝑇U) = 1 2 [1 + erf( 𝑇U − 𝜇 √2 ⋅ σ )] , (10) while 𝐹𝑋(𝑇L) can be similarly obtained. In the above equation, 𝜇, the mean value of the normal PDF, represents the measured value 𝑥m of the measurand. Therefore, if we want to find AU (AL), that is the maximum (minimum) value of the measured value such that 𝑝w ≤ MAR, 𝜇 = 𝐴U (𝜇 = 𝐴L) must be considered in (10). Therefore, according to (5) and (6) 1 2 [1 + erf( 𝑇U,L − 𝐴U,L √2 σ )] ={ 1 − MAR if 𝑥m < 𝑇U is required MAR if 𝑥m > 𝑇L is required. (11) When 𝑥m < 𝑇U is required, solving equation (11) yields: 𝐴U = 𝑇U − √2 σ ⋅ erfinv(1 − 2 ⋅ MAR) , (12) where erfinv is the inverse error function, which is given by: erfinv(𝑧) = ∑ 𝑐𝑘 2 𝑘 + 1 ∞ 𝑘=0 ( √π 2 𝑧) 2 𝑘+1 , (13) where 𝑐0 = 1 and 𝑐𝑘 = ∑ 𝑐𝑚 𝑐𝑘−1−𝑚 (𝑚 + 1)(2𝑚 + 1) 𝑘−1 𝑚=0 . Similarly to the error function, also the inverse error function is well approximated with no more than 10 terms in (13). On the other hand, when 𝑥m > 𝑇L is required, solving (11) yields: 𝐴L = 𝑇𝐿 − √2 σ ⋅ erfinv(2 ⋅ MAR − 1) . (14) Since the inverse error function is an anti-symmetric function, that is: erfinv(−𝑧) = − erfinv(𝑧) , (15) equations (12) and (14) can be grouped into a single equation: 𝐴U,L = 𝑇U,L ∓ √2 σ ⋅ erfinv(1 − 2 ⋅ MAR) . (16) Therefore, to have a risk lower than MAR to exceed the tolerance limit TU (or TL), an acceptance limit AU (or AL) should be evaluated, obtained by shifting the tolerance limit to the left (or right) by quantity √2 ⋅ σ ⋅ erfinv(1 − 2 ∙ 𝑀𝐴𝑅). In particular: • the limit is shifted to the left when 𝑥𝑚 ≤ 𝑇U is required and guarded acceptance is applied; • the limit is shifted to the right when 𝑥𝑚 ≥ 𝑇L is required and guarded acceptance is applied; • the limit is shifted to the right when 𝑥𝑚 ≤ 𝑇U is required and guarded rejection is applied; • the limit is shifted to the left when 𝑥𝑚 ≥ 𝑇L is required and guarded rejection is applied. To provide a numerical example, let us consider again the example considered in Section 2.2 and the values in Table 1. Let us remember that 𝑥𝑚 < 𝑇U is required and suppose the MAR is set to 5 % and guarded acceptance is considered. By applying equation (16), it follows 𝐴U = 41.8 mg/l. Figure 6 shows the normal PDF with a mean value equal to the obtained AU and standard uncertainty given in Table 1. In this figure, the coloured area represents the probability of being above TU, as also reported in Table 2. This probability is exactly the set MAR. This means that, for every value 𝑥𝑚 < 𝐴U, the probability of exceeding 𝑇U will be lower than 5%. It is therefore possible to set the acceptance limit, given the PDF associated with the estimated measurement uncertainty and the desired MAR. 3.2. The measurement results distribute according to a uniform posterior PDF When a uniform PDF is considered: 𝑝(𝑥) = { 1 2𝑎 if 𝜇 − 𝑎 < 𝑥 < 𝜇 + 𝑎 0 otherwise , (17) where 𝜇 is the mean value and 2a is the support of the PDF, which is related to the PDF standard deviation by 𝑎 = 𝜎 ⋅ √3. The corresponding CDF is given by: 𝐹𝑋(𝑥) = ∫ 𝑝(𝑡) d𝑡 = 𝑥 −∞ ∫ 1 2 𝑎 d𝑡 = 𝑥 𝜇−𝑎 1 2 𝑎 .(𝑥 − 𝜇 + 𝑎) (18) Figure 6. Example when the normal PDF is centered on AU = 41.8 mg/l. The coloured area represents the probability of being above TU. Table 2. Probability of being below or above TU in the case of Figure 6. 𝑷(𝒙 < MAL) 𝑷(𝒙 > MAL) 0.95 0.05 ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 6 and therefore: 𝐹𝑋(𝑇U,L) = 1 2 𝑎 ⋅ (𝑇U,L − 𝜇 + 𝑎) . (19) From (5) and (6), and considering 𝜇 = 𝐴U,L in (19): 1 2a ⋅ (𝑇U,L − 𝐴U,L + a) = = { 1 − MAR if 𝑥𝑚 < 𝑇U is required MAR if 𝑥𝑚 > 𝑇𝐿 is required . (20) By solving the above equations, the value for the acceptance limit is found: 𝐴U,L = 𝑇U,L ∓ 𝑎 ⋅ (1 − 2 ⋅ MAR) (21) which shows that, to have a risk below MAR, the acceptance limit must be set to the left or right of the tolerance limit (as detailed in the previou section) by quantity 𝑎 ⋅ (1 − 2 ⋅ MAR). To provide a numerical example, let us consider again the example of the pollutant in water considered in Section 2.2. Let us consider again that TU = 50 mg/l as in Table 1, but let us now suppose that the PDF associated with the estimated measurement uncertainty is uniform. Let us also suppose that the half-width of this uniform PDF is a = 10 mg/l, that the MAR is set to 5 % and that guarded acceptance is applied. By applying equation (21), it follows AU = 41 mg/l. Figure 7 shows the uniform PDF with a mean value equal to the obtained AU. The coloured area represents the probability of being above TU, which is exactly 5 %. This means that every measured value of the pollutant in water lower than AU will provide a risk of exceeding TU lower than 5 %. 3.3. The measurement results distribute according to a triangular posterior PDF When a symmetric triangular PDF is considered, its equation is the following: 𝑝(𝑥) = { 𝑦1(𝑥) if 𝜇 − 𝑎 ≤ 𝑥 ≤ 𝜇 𝑦2(𝑥) if 𝜇 < 𝑥 ≤ 𝜇 + 𝑎 0 otherwise, (22) where: 𝑦1(𝑥) = 𝑥 𝑎2 + 𝑎 − 𝜇 𝑎2 (23) and: 𝑦2(𝑥) = − 𝑥 𝑎2 + 𝑎 + 𝜇 𝑎2 (24) and where 𝜇 and 2a are, respectively, the mean value and the support of the PDF. Furthermore, 𝑎 = 𝜎 √6 holds, where 𝜎 is the standard deviation of the PDF. To evaluate the corresponding CDF, two situations should be considered, that is the case when 𝑥 ≤ μ and the case when 𝑥 > 𝜇. If 𝑥 ≤ μ, then the CDF is given by: 𝐹𝑋,1(𝑥) = ∫ 𝑦1(𝑡) d𝑡 𝑥 𝜇−𝑎 = ∫ ( 𝑡 𝑎2 + 𝑎 − 𝜇 𝑎2 )d𝑡 𝑥 𝜇−𝑎 = 1 2 𝑎2 ⋅ [𝑥2 + 2𝑥 ⋅ (𝑎 − 𝜇) + (𝑎 − 𝜇)2] = 1 2 𝑎2 ⋅ [𝑥 + (𝑎 − 𝜇)]2 , (25) while, if 𝑥 > μ, the CDF is given by: 𝐹𝑋,2(𝑥) = 1 2 + ∫ 𝑦2(𝑡) d𝑡 = 1 2 𝑥 𝜇 + ∫ (− t a2 + a + 𝜇 a2 ) 𝑥 𝜇 d𝑡 = 1 2 − 1 2 𝑎2 . [𝑥2 − 2 𝑥 ∙ (𝑎 + 𝜇) + 𝜇 ∙ (𝜇 + 2𝑎)] . (26) Now, (5) and (6) should be solved for both 𝐹𝑋,1 and 𝐹𝑋,2, thus leading to four equations. However, only the most likely situations are here reported. In fact, when 𝑥 ≤ 𝑇U is required and the MAR is supposed to be small (as it should be when environmental, legal or health situations are considered), the situation shown in Figure 8a will occur, so that equation (27) must be solved. 𝐴U|𝐹𝑋,2(𝑇U) = 1 − MAR . (27) On the other hand, when 𝑥𝑚 ≥ 𝑇L is required and again the MAR is supposed to be small, the situation in Figure 9b will occur, so that equation (28) must be solved. 𝐴L|𝐹𝑋,1(𝑇L) = MAR . (28) Equation (27) yields: 1 2 − 1 2 𝑎2 ⋅ [𝑇U 2 − 2 ∙ 𝑇U ⋅ (𝑎 + 𝐴U) + 𝐴U ⋅ (𝐴U + 2𝑎)] = 1 − MAR . (29) By solving this simple equation with respect to AU, the following second-order equation is found: 𝐴U 2 − 2 ∙ 𝐴U ⋅ (𝑇U − 𝑎) + [(𝑇U − 𝑎) 2 − 2 ∙ 𝑎2 ⋅ MAR] = 0 , (30) Figure 7. Example when the uniform PDF is centered on AU = 41 mg/l. The coloured area represents the probability of being above TU. Figure 8. Example when a triangular PDF is assumed. The coloured area represents the probability of exceeding TU (or TL) when: a) the measured value is required to be below the tolerance limit (𝑥𝑚 ≤ 𝑇U); b) the measured value is required to be above the tolerance limit (𝑥𝑚 ≥ 𝑇L). ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 7 which provides the two solutions: 𝐴U = (𝑇U − 𝑎) ∓ 𝑎 ∙ √2 ∙ MAR . (31) Among the two above solutions, the one with the minus sign can be discarded. In fact, if we considered a PDF with width 2a and 𝜇 = (𝑇U − 𝑎) − 𝑎 ∙ √2 ∙ MAR, this PDF would not cross TU and therefore it would provide a risk of exceeding TU equal to zero. Of course, this should be a very lucky situation, but here the limit not to exceed the MAR needs to be found and, therefore, the following equation holds, under the assumption that 𝑥 ≤ 𝑇U is required: 𝐴U = (𝑇U − 𝑎) + 𝑎 ∙ √2 ∙ MAR . (32) When, on the other hand, 𝑥 ≥ 𝑇L is required, (28) yields: 1 2 𝑎2 ⋅ [𝑇L 2 + 2 ∙ 𝑇𝐿 ⋅ (𝑎 − 𝐴L) + (𝑎 − 𝐴L) 2] = MAR . (33) Solving this simple equation with respect to AL, the following second-order equation is found: 𝐴L 2 − 2 ∙ 𝐴L ⋅ (𝑇L + 𝑎) + [(𝑇L + 𝑎) 2 − 2 ∙ 𝑎2 ⋅ MAR] = 0 , (34) which provides the two solutions: 𝐴L = (𝑇L + 𝑎) ∓ 𝑎 ∙ √2 ∙ MAR . (35) Among the two above solutions, the one with the plus sign can be discarded. In fact, if we considered a PDF with width 2a and 𝜇 = (𝑇L + 𝑎) + 𝑎 ∙ √2 ∙ MAR, this PDF would not cross TL and therefore it would provide a risk of exceeding TL equal to zero. Of course, this should be a very lucky situation, but here the limit not to exceed the MAR needs to be found and, therefore, the following equation holds, under the assumption that 𝑥 ≥ 𝑇L is required: 𝐴𝐿 = (𝑇L + 𝑎) − 𝑎 ∙ √2 ∙ MAR . (36) Finally, by considering (32) and (36) together, it is: 𝐴U,L = 𝑇𝑈,𝐿 ∓ 𝑎 ⋅ (1 − √2 ∙ MAR) , (37) that is, the acceptance limit must be shifted to the left or right of the tolerance limit (as detailed in Sec. 3.1) by quantity 𝑎 ⋅ (1 − √2 ∙ MAR) to have a risk lower than the MAR. As a numerical example, let us consider again the example of the pollutant in water considered in Section 2.2. Let us consider again that TU = 50 mg/l as in Table 1, but let us now assume a triangular PDF associated to the estimated uncertainty. Furthermore, the half-width of the triangular PDF is supposed to be a = 10 mg/l, the MAR is set to 5% and guarded acceptance is applied. Since 𝑥 ≤ 𝑇U is desired, by applying (37), AU = 43.2 mg/l is obtained. Figure 9 shows the obtained PDF (centered on the obtained AU value), where the coloured area represents the probability of being above the tolerance limit TU, which is exactly equal to the pre-set MAR (5%). This means that every measured value of the pollutant in water lower than the obtained AU value will provide a risk lower than 5 %. 3.4. The measurement results distribute according to a trapezoidal posterior PDF When a symmetric trapezoidal PDF is considered, then the PDF is described by the following equations: 𝑝(𝑥) = { y3(x) if μ − 𝑎 ≤ 𝑥 ≤ μ − 𝑎 β 1 𝑎 ⋅ (1 + β) if μ− 𝑎 β < 𝑥 ≤ μ + 𝑎 β 𝑦4(𝑥) if   μ + 𝑎 β < 𝑥 ≤ μ + 𝑎 0 otherwise, (38) where: 𝑦3(𝑥) = 1 𝑎2 ⋅ (1 − β2) ⋅ (𝑥 + 𝑎 − 𝜇)  (39) 𝑦4(𝑥) = − 1 𝑎2 ⋅ (1 − β2) ⋅ (𝑥 − 𝑎 − 𝜇) , (40) 𝜇 is the mean value of the PDF, 2a is its width and  is the ratio between the two basis. To evaluate the corresponding CDF, three situations should be considered, that is the case when 𝜇 − 𝑎 ≤ 𝑥 ≤ 𝜇 − 𝑎 𝛽, the case when 𝜇 − 𝑎 𝛽 < 𝑥 ≤ 𝜇 + 𝑎 𝛽 and the case when 𝜇 + 𝑎 𝛽 < 𝑥 ≤ 𝜇 + 𝑎. However, similar considerations as the ones made for the triangular PDF apply, so that only the two situations shown in Figure 10a and Figure 10b are considered. Let us call 𝐹𝑋,3(𝑥) the CDF for the case in Figure 10a. The following equation must then be solved: 𝐴𝑈|𝐹𝑥,3(𝑇U) = 1 – MAR . (41) On the other hand, let us call 𝐹𝑋,4(𝑥) the CDF for the case in Figure 10b. The following equation must then be solved: 𝐴𝐿|𝐹𝑋,4(𝑇𝐿) = MAR . (42) Figure 9. Example when the triangular PDF is centered on AU = 43.2 mg/l. The coloured area represents the probability of being above TU. Figure 10. Example when a trapezoidal PDF is assumed. The coloured area represents the probability of exceeding the tolerance limit when: a) the measured value is required to be below the tolerance limit (𝑥𝑚 ≤ 𝑇U); b) the measured value is required to be above the tolerance limit (𝑥𝑚 ≥ 𝑇𝐿). ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 8 It follows: 𝐹𝑥,3(𝑥) = 1 − 𝛽 2 (1 + 𝛽) + 2 𝛽 1 + 𝛽 + ∫ 𝑦4(𝑡) d𝑡 𝑥 𝜇+𝑎𝛽 = 1 + 3 𝛽 2 (1 + 𝛽) + 1 2 𝑎2(1 − 𝛽2) ∙ ⋅ {−𝑥2 + 2(𝜇 + 𝑎) ⋅ [𝑥 − (𝜇 + 𝑎 𝛽)]+(𝜇 + 𝑎 𝛽)2} (43) and, by solving (41), the second-order equation is obtained: 𝐴U 2 − 2 ∙ 𝐴U ⋅ (𝑇U − 𝑎) + (𝑇U − 𝑎) 2 − 2 ∙ 𝑎2 ⋅ MAR ⋅ (1 − 𝛽2) = 0, (44) for which the two following solutions can be found: 𝐴U = (𝑇U − 𝑎) ∓ 𝑎 ⋅ √2 ∙ MAR ⋅ (1 − 𝛽 2) . (45) According to the same considerations as the ones in the case of a triangular PDF, it can be concluded that, among the two above solutions, the one with the minus sign can be discarded. Therefore, the following equation holds, under the assumption that x  TU is required: 𝐴U = (𝑇U − 𝑎) + 𝑎 ⋅ √2 ∙ MAR ⋅ (1 − 𝛽 2) . (46) On the other hand: FX,4(x)=∫ y3(t) dt x μ-a = 1 2 𝑎2(1-𝛽2) ⋅[x2+2∙(a-μ)∙𝑥+(𝑎 − 𝜇)2] (47) and, by solving (42) with respect to AL, the following second- order equation is obtained: 𝐴L 2 − 2 ∙ 𝐴L ⋅ (𝑇L + 𝑎) + (𝑇L + 𝑎) 2 − 2 ∙ 𝑎2 ⋅ MAR ⋅ (1 − 𝛽2) = 0 (48) Equation (48) has two solutions: 𝐴L = (𝑇L + 𝑎) ∓ 𝑎 ⋅ √2 ∙ MAR ⋅ (1 − 𝛽 2) , (49) where, according to the same previous considerations, the one with the plus sign can be discarded, so that: 𝐴L = (𝑇L + 𝑎) − 𝑎 ⋅ √2 ∙ MAR ⋅ (1 − 𝛽 2) . (50) Finally, by considering together (46) and (50), it can be written: 𝐴U,L = 𝑇U,L ∓ 𝑎 ⋅ (1 − √2 ∙ MAR ⋅ (1 − 𝛽 2)) , (51) that is, the acceptance limit is simply shifted to the left or right of the tolerance limit (as detailed in Sec. 3.1) by quantity 𝑎 ⋅ (1 − √2 ∙ MAR ⋅ (1 − 𝛽2)). To provide a numerical example, let us consider again the example of the pollutant in water considered in Section 2.2. Let us consider again that TU = 50 mg/l as in Table 1, but let us now suppose that the PDF is trapezoidal with  = 0.5. Furthermore, the half-width of the trapezoidal PDF is supposed to be 𝑎 = 10 mg/l, the MAR is set to 5 % and guarded acceptance is applied. Since, in the considered example, x  TU is required, (51) yields AU = 42.7 mg/l. Figure 11 shows the obtained PDF (centered on the obtained AU value), where the coloured area represents the probability of being above the tolerance limit TU and it is exactly equal to the pre-set MAR (5 %). This means that every measured value of the pollutant in water lower than the obtained AU value will provide a risk lower than 5 %. 4. CONCLUSIONS Following the suggestions given in the present Standards [14] [16], a measurand is considered as conforming when the measured value falls inside the acceptance interval, as defined in art. 3.3.9 of [14], and is considered non-conforming when the measured value falls inside the rejection interval, as defined in art. 3.3.10 of [14]. The definition of the acceptance interval strongly depends on the measurement uncertainty with which the measurand is measured and the admissible risk of declaring a non-conforming value as conforming and vice versa. While this problem is clearly highlighted in [14], very few practical indications are given on how to define the acceptance limits once the measurement uncertainty is estimated, and the maximum admissible risk (MAR) given. This paper has shown how the acceptance limits depend also on the probability density function (PDF) considered to represent the distribution of values that could reasonably be attributed to the measurand, and has proposed a general method to relate them to the considered PDF and the considered MAR. The most used normal, uniform, triangular and trapezoidal PDFs have been considered and general formulas have been given to define the acceptance limits given uncertainty and MAR. The numerical examples have shown that different results are obtained for the acceptance limits, when the different PDFs are considered, as expected from the theory. The closed-form formulas provided in the paper allow one to evaluate the acceptance limits in a straightforward way, in both situations of guarded acceptance and guarded rejection. Should different probability distributions be considered, the general proposed method can still be applied, and a Monte Carlo simulation can provide the desired acceptance limits. REFERENCES [1] JCGM 200:2012. International Vocabulary of Metrology – Basic and General Concepts and Associated Terms (VIM 2008 with minor corrections). Joint Committee for Guides in Metrology. 2012. [2] JCGM 100:2008. Evaluation of Measurement Data – Guide to the Expression of Uncertainty in Measurement, (GUM 1995 with minor corrections). Joint Committee for Guides in Metrology. 2008. Figure 11. Example when the trapezoidal PDF is centered on AU = 42.7 mg/l.. The coloured area represents the probability of being above TU. ACTA IMEKO | www.imeko.org June 2023 | Volume 12 | Number 2 | 9 [3] A. Ferrero, V. Scotti. Uncertainty and Conscious Decisions, In: Forensic Metrology: An Introduction to the Fundamentals of Metrology for Judges, Lawyers and Forensic Scientists. Cham: Springer International Publishing, 2022, pp. 115–124. ISBN: 978- 3-031-14619-0. DOI: 10.1007/978-3-031-14619-0_8 [4] M Peterson, An Introduction to Decision Theory. Cambridge: Cambridge University Press, 2009, p. 317. ISBN: 9780511800917. DOI: 10.1017/CBO9780511800917 [5] C Yoe, Principles of Risk Analysis: Decision Making Under Uncertainty. 2nd ed. Boca Raton: CRC Press, 2019, p. 848. ISBN: 9780429021121. DOI: 10.1201/9780429021121 [6] R. M. Peterman, J. L. Anderson, Decision Analysis: A Method for Taking Uncertainties into Account in Risk- Based Decision Making, Human and Ecological Risk Assessment: An Int. Journal 5.2 (1999), pp. 231–244. DOI: 10.1080/10807039991289383 [7] B. Solaiman, D. Guériot, Sh. Almouahed, B. Alsahwa, É. Bossé, A New Hybrid Possibilistic-Probabilistic Decision-Making Scheme for Classification, Entropy 23.1 (2021). ISSN: 1099-4300. DOI: 10.3390/e23010067 [8] L. R. Pendrill, Using measurement uncertainty in decision-making and conformity assessment, Metrologia 51.4 (2014), pp. 206–218. DOI: 10.1088/0026-1394/51/4/S206 [9] A. Allard, N. Fischer, I. Smith, P. Harris, L. Pendrill, Risk calculations for conformity assessment in practice. Ed. by Array, 19th International Congress of Metrology CIM, Paris, France, 24- 26 September 2019. DOI: 10.1051/metrology/201916001 [10] S. Puydarrieux, J. M. Pou, L. Leblond, N. Fischer, A. Allard, M. Feinberg, D. El Guennouni, Role of measurement uncertainty in conformity assessment. Ed. by Array, 19th International Congress of Metrology CIM, Paris, France, 24-26 September 2019. DOI: 10.1051/metrology/201916003 [11] E. Cruz de Oliveira, F. R. Lourenço, Risk of false conformity assessment applied to automotive fuel analysis: A multiparameter approach, Chemosphere 263 (2021), p. 128265. ISSN: 0045-6535. DOI: 10.1016/j.chemosphere.2020.128265 [12] L. Separovic, R. S. Simabukuro, A. R. Couto, M. L. G. Bertanha, F. R.S. Dias, A. Y. Sano, A. M. Caffaro& F. R. Lourenço, Measurement Uncertainty and Conformity Assessment Applied to Drug and Medicine Analyses – A Review, Critical Reviews in Analytical Chemistry 0.0 (2021), pp. 1–16. DOI: 10.1080/10408347.2021.1940086 [13] M. H. Habibie, O. Hedrony, Using decision rule in calibration of long gauge block on the conformity assessment scheme, IOP Conference Series: Materials Science and Engineering 673.1 (Dec. 2019), p. 012110. DOI: 10.1088/1757-899X/673/1/012110 [14] JCGM 106:2012. Evaluation of measurement data – The role of measurement uncertainty in conformity assessment. Joint Committee for Guides in Metrology. 2012. [15] J. M. Pou, L. Leblond, Smart Metrology: From the metrology of instrumentation to the metrology of decisions. Ed. by Array, 18th International Congress of Metrology CIM, Paris, France, 19-21 September 2017. DOI: 10.1051/metrology/201701007 [16] ISO EN 14253-1. Geometrical Product Specification (GPS) - Inspection by measurement of workpieces and measurement instruments - Part 1: Decision rules for proving conformance or non-conformance with specification. ISO EN 14253-1. 1998 M [17] H. Källgren, L. Pendrill, B. Magnusson. Role of measurement uncertainty in conformity assessment in legal metrology and trade, Accreditation and Quality Assurance 8 (2003), pp. 541–547. DOI: 10.1007/s00769-003-0707-8 [18] JCGM 101:2008. Evaluation of measurement data – Supplement 1 to the Guide to the expression of uncertainty in measurement – Propagation of distributions using a Monte Carlo method. Joint Committee for Guides in Metrology. 2008. https://doi.org/10.1007/978-3-031-14619-0_8 https://doi.org/10.1017/CBO9780511800917 https://doi.org/10.1201/9780429021121 https://doi.org/10.1080/10807039991289383 https://doi.org/10.3390/e23010067 https://doi.org/10.1088/0026-1394/51/4/S206 https://doi.org/10.1051/metrology/201916001 https://doi.org/10.1051/metrology/201916003 https://doi.org/10.1016/j.chemosphere.2020.128265 https://doi.org/10.1080/10408347.2021.1940086 https://dx.doi.org/10.1088/1757-899X/673/1/012110 https://doi.org/10.1051/metrology/201701007 https://doi.org/10.1007/s00769-003-0707-8