1. Introduction
Seeking accreditation, synonymous with technical competence and reliability, should be a primary strategic decision for any calibration or testing laboratory. Implementing a quality management system based on the criteria of the ISO/IEC 17025:2017 [
1] standard is crucial, as it acknowledges the validity and certainty of results issued by testing laboratories for conformity assessment purposes.
In this context, the Mexican Accreditation Entity (EMA) requires a set of requirements to attain accreditation, in addition to those outlined in the ISO/IEC 17025:2017 [
1] standard. These requirements are detailed in the document “Application Criteria of the ISO standard 17025 (MP-FE005-13)”. Through evaluation and practical experience, the accrediting body has updated the application criteria of ISO 17025:2017 to clarify and interpret the management and technical requirements of the standard that are deemed critical.
Moreover, this document aims to establish complementary requirements that form part of the evaluation criteria used in the accreditation processes of testing and calibration laboratories. These criteria are essential for laboratories, dependencies, accredited laboratories, members of the national registry of evaluators, and interest groups to adhere to, ensuring clarity and interpretation of the ISO/IEC 17025:2017 standard’s management and technical requirements.
The strategy for accrediting any test or calibration method, as outlined in the preceding paragraph, is subject to a dual set of criteria: the primary criteria of the ISO/IEC 17025:2017 [
1] standard and an additional set of interpretation criteria provided by the EMA in the document MP-FE005-13. Within this framework, Section 7.2 of the ISO/IEC 17025:2017 standard delineates the requirements, both implicit and mandatory, for validating any test, trial, or calibration method. This section, specifically titled “Selection, Verification, and Validation of Methods,” is crucial for ensuring the accuracy and reliability of the methods employed.
It should be understood that both references recognize two types of methods: regulated ones, which undergo verification, and non-regulated or proprietary methods, which require validation. While the ISO/IEC 17025:2017 [
1] standard mandates scope validation for methods, the EMA MP-FE005-13 guide provides a more detailed description of the criteria to be addressed for these test methods. The specific items mentioned above will be further discussed in subsequent sections of this paper.
Following a discussion on the validation process for non-regulated methods, it is essential to provide a clear description of the method under consideration and outline the necessary documentation to cover the validation requirements outlined previously. The method in question, colloquially known in Mexican mines as the ‘anchor test’, formally titled ‘Strength Determination in Anchors Installed Inside the Mine’, will be expounded in the following section.
The method “Resistance Determination in Anchors Installed Inside the Mine” is not a recent innovation but rather an established process rooted in the principle of material adherence. Specifically, it focuses on the bond between certain metals and mortar. This fundamental phenomenon is crucial for the method’s functionality. As highlighted by Cedeño and González [
2], without this adherence, steel bars would lack the ability to withstand even minimal tensile stress, as the steel would simply slide along its entire length without resistance. However, due to the presence of adherence, stress is evenly distributed throughout the material, allowing the steel to bear the load. Even when cracks occur, they are distributed evenly, maintaining the bond between the materials in the cracked areas.
The adhesion serves two purposes: first, it ensures anchorage, primarily provided by the length of the steel barrel; second, it transmits peripheral tangential forces arising in the main reinforcement due to variations in longitudinal stress. This phenomenon of adhesion is attributed to the physical–chemical nature, facilitated by the capillary and molecular forces developed at the interface between steel and concrete. Here, the steel absorbs cement paste, aided by the retraction effect [
2].
Adhesion failure occurs when the pulling force is transmitted to the concrete via the ridges or cords of the metal bar, creating an internal stress in a ring-shaped manner. This stress leads to cracks along the embedded metal bar. As the load on these rings increases until failure, longitudinal cracks emerge, ultimately causing the cone to fail at an angle of inclination, denoted as alpha. These cracks exacerbate the slippage between the bar and the mortar. Once the radial component reaches its peak, the mortar abruptly fractures [
3].
Figure 1 illustrates this type of failure.
One of the earliest studies on the adhesion phenomenon, referenced in the preceding paragraphs, dates back to the year 1899. This study involved measuring the displacement of the ends of steel cables embedded in mortar, followed by subjecting them to tensile loading. From this test, a force/displacement relationship was established, which became the standard method for measuring adhesion until the 1950s. This marked the inception of structural adhesion studies for materials.
In 1958, Klaus Müller Rehm publicly disclosed several results from an adhesion experiment, marking a significant development in the field. This experiment introduced the concrete rod extraction test, commonly known as “The Pull Out Test.” The experiment involved immersing a specially shaped steel bar into concrete, ensuring it was embedded within the steel body. Subsequently, the bar was subjected to tensile stress.
Figure 2 illustrates this experimental setup.
The results of the previous experiment led to the identification of two types of adhesion failures. In the first type, failure occurs due to bond breaking, characterized by the extraction of the steel bar accompanied by radial cracks in the concrete, emanating from the surface of the bar. In the second type of failure, damage to the concrete occurs due to the development of high internal stress, resulting in the appearance of longitudinal cracks parallel to the axis of the bar and spreading in the same direction [
4].
In 1979, Ralejs Tepfers (1933–2015) pioneered one of the earliest formal scientific investigations into determining the resistance of metal bars subjected to tensile stress and the interaction between these elements. In this seminal work, the experimental interface was analytically modeled as a layer where both shear and internal pressure simultaneously play a role. As a result, it was proposed that the resistance of this interaction is determined by the ability of the surrounding concrete or mortar to withstand the circumferential stress induced by the steel bar [
4]. This marked the inception of what is now known as the “Determination of Strength in Anchors” test.
According to [
5], the validation of test methods aims to ensure that the specified requirements are suitable for their intended use. During the validation process, the laboratory verifies the method’s appropriateness for use, tailoring the validation extent to meet the application’s needs. The methods employed for validation can include one or a combination of the following:
Measurement and precision measurement using standard samples.
Systematic evaluation of factors influencing the results.
Testing the method’s stability by varying regulated parameters such as incubator temperature or dose volume.
Comparing results with those obtained from other validated methods.
Conducting inter-laboratory comparisons.
Estimating the uncertainty of measurement results based on an understanding of the method’s theoretical principles and practical experience with the sampling or test method.
These validation methods ensure that test methods are reliable, providing accurate and consistent results.
Similarly, [
6] outlines common validation methods in their study, including the following:
Robustness studies, which assess the test method’s ability to yield consistent results under different conditions.
Precision and accuracy studies, aimed at determining the precision and accuracy of results obtained through the test method.
Repeatability and reproducibility studies, which evaluate result variability when tests are repeated under identical conditions and between different operators.
Building upon [
7], according to ISO/IEC 17025, validation of a test method confirms, through objective evidence, its suitability for its intended use. Verification, on the other hand, confirms, also through objective evidence, that the test method is suitable for its specific application. Laboratories typically conduct validation experiments, inter-laboratory comparisons, uncertainty analyses, and other technical procedures to demonstrate the competence and reliability of the methods used in testing and calibration.
The three studies converge on recommending three precision methods: r&R studies, precision and accuracy studies and inter-laboratory testing. Despite thorough research, no studies or articles on the validation of anchoring methods were found at the time of the investigation. However, the Government of Canada conducted a comparative test on anchor resistance, categorizing the results based on design type rather than according to the extraction procedure [
8].
The objective of the present study is to investigate the implementation of ISO 17025 accreditation standards in testing laboratories engaged in the evaluation of rock support systems. It aims to analyze the accreditation’s impact on the reliability and traceability of test results, as well as on the quality assurance processes within rock engineering practices. This study seeks to establish standardized testing protocols aligning with ISO 17025 requirements for evaluating the performance of bolts, cables, and anchors, ensuring the accuracy and comparability of results across different laboratories. Additionally, it intends to explore the integration of quality management systems based on ISO 17025 principles in the design and execution of rock support projects, with the goal of enhancing safety, efficiency, and compliance with international standards in the rock engineering industry.
Additionally, the aim is to propose a procedure for validating the test method before any accreditation body, drawing from a case example in Mexico. This proposal aims to enhance safety, efficiency, and compliance with international standards in the rock engineering industry.
2. Materials and Methods
Starting from the premise that adhesion is the property of materials whereby two surfaces of different substances are held together by intermolecular forces upon contact, it becomes evident that adhesion is a fundamental phenomenon underlying the effectiveness of indoor shotcrete. In mining, adhesion plays a critical role in the structural integrity of the material, particularly in the bond between mortar and steel (steel bars anchored directly into rock and submerged in mortar). Without such adhesion, steel bars would fail to withstand even minimal tensile stress, as the steel would simply slide along its entire length without encountering resistance. However, due to adhesion, when cracks occur, they distribute evenly throughout the material, enabling the steel to withstand tensile stress and maintaining the bond between the materials in cracked areas. This supports the prevention of potential rock detachment and ensures the continuity of mining operations with maximum safety. Therefore, anchors remain firmly placed in pre-drilled holes in the rock, combining rock and soil.
Under these conditions, the space between the hole and the anchor is filled with mortar. In this configuration, the anchor operates passively, relying on friction, as illustrated in
Figure 3.
Both soil and rock are natural materials that exhibit a wide range of engineering properties, varying significantly from one location to another. The capacity of an anchor is influenced by its depth of placement. Rock anchors typically have medium to very high capacity, as rock can withstand higher stress concentrations. The hole typically maintains a constant cross-section. The shear strength of the anchor is generally less influenced by the depth of anchorage [
10].
The anchor installation procedure typically involves three steps:
- (a)
Rock drilling: Holes are drilled for the anchor with diameters and depths defined by the project. Pneumatic or rotary equipment is typically used for drilling.
- (b)
Hole cleaning: once drilling is complete, the hole is cleaned by either blowing air or washing it with water until all drilling waste material is removed.
- (c)
Mortar placement: Subsequently, mortar is poured into the perforation using a hose of sufficient length. The mortar is deposited at the bottom of the hole first, then advanced upwards, with the hose always submerged in the mixture [
11]. The installed anchor can be seen in
Figure 4.
Once the study element has been described, it is crucial to recall the requirements stipulated by the EMA for the validation of the proposed method (as outlined in
Table 1).
These requirements apply specifically to tests involving physical measurements:
Verification of equipment performance against the method’s established requirements.
Consistency of laboratory facilities and environmental conditions with those specified in the method.
Conducting repeatability and reproducibility studies to ensure compliance with acceptance criteria based on performance data.
Providing evidence of inter-laboratory comparisons.
These points establish the guidelines to be developed in this methodological section. Meeting these requirements is essential for validating a non-regulated or proprietary test method.
Load Capacity: The load capacity of an anchor bolt is crucial as it determines its effectiveness in controlling underground instability. It is essential to differentiate between the support element before insertion and the one applied to the rock mass (the anchor). The anchor’s load capacity is measured once it is placed within the rock mass. Therefore, it is important to understand that the concept of anchorage arises from the interaction between the rock mass, the steel bolt, and the applied encapsulant mix (mortar/concrete/resin), such as cement cartridges, resin cartridges, and cement grout injections.
Equipment elements:
- (a)
Equipment castle;
- (b)
Hydraulic cylinder;
- (c)
Safety chain;
- (d)
High-pressure hose;
- (e)
Pressure indicator gauge;
- (f)
Hydraulic pump;
- (g)
Butterfly-type safety nut.
These equipment elements are crucial for accurately measuring and verifying the load capacity of the anchor bolt in accordance with the method’s requirements.
It is crucial to note that to fulfill this initial section, the equipment, pressure gauge, and hydraulic pump must undergo calibration by an accredited calibration provider. The calibration interval should align with the anchor performance range during the pull strength test, which typically falls between 1 and 15 tons. Additionally, it is essential to verify that the calibration remains valid on the test date. By meeting these requirements, there is ample evidence to verify the equipment’s performance according to the criteria established in the method.
- 2.
Agreement on Laboratory Facilities and Environmental Conditions with Method Specifications:
The actual load capacities provided by the anchor have been documented based on the characteristic geomechanical conditions of the reservoir (environmental conditions) and design requirements. These activities must be conducted for each characteristic geomechanical domain of the underground project, distinguishing between dominant and random domains in each underground area [
12].
Regarding laboratory facilities, testing is conducted at prevailing locations, always prioritizing the most accurate security measures. It is essential to ensure that the laboratory facilities and environmental conditions align with the specifications outlined in the method to maintain consistency and reliability in testing procedures.
- 3.
Repeatability and Reproducibility (r&R) Study in Compliance with Acceptance Criteria through Analysis of Performance Data:
Errors in any measurement system, and consequently in any test, stem from factors related to accuracy (the difference between the measured value and the true value of an item) or precision (variations observed when the same item is repeatedly measured with the same device). Similarly, the accuracy of a measurement system comprises components of variation associated with the measurement instrument, which can be minimized or reduced to an acceptable extent through proper calibration within required intervals.
The accuracy of a measurement system consists of two main components:
Repeatability: variation can be caused by the measurement device and is observed when the same operator measures the same item multiple times using the same measurement system under identical conditions.
Reproducibility: variation can be caused by the measurement system and is observed when different operators measure the same item multiple times using the same measurement system under consistent conditions.
Conducting a repeatability and reproducibility study helps assess the performance of the measurement system and ensures that it meets acceptance criteria through an analysis of performance data.
Therefore, to evaluate the precision of a measurement system, it is recommended to utilize r&R (repeatability and reproducibility) studies. These studies measure the error generated by the two components mentioned earlier and also provide insight into the intrinsic variability within the measurement system, which can be challenging to control. In essence, r&R studies compare the variation in the measurement system components with their total variation and tolerance.
The methods employed to conduct r&R studies include the following:
- (a)
Range method (also known as the short method);
- (b)
Average and range method (also known as the long method);
- (c)
Variance analysis method (also known as the Anova method).
Indeed, the variance analysis method presents several advantages compared to the first two methods. Firstly, it offers a more accurate estimation of variances, providing a better understanding of the variability within the measurement system. Additionally, it yields valuable additional information, such as identifying operator interaction with the item being measured. Moreover, the variance analysis method considers a random effects model for all factors or components of the system’s variation, including operators, items, interaction, and instrument. This comprehensive approach ensures a thorough assessment of the measurement system’s performance and helps in identifying sources of variation that may affect the accuracy and reliability of the measurements.
The standard ANOVA table, which includes the formulas for the sums and mean squares of each system variation component, is presented in
Table 2 [
13].
Derived from the fact that the anchor test installed inside the mine is a destructive test, the same sample cannot be measured multiple times. Therefore, it was decided to perform tests at different points while maintaining consistency in the supplier of the anchor rods. This approach aims to reduce potential sources of variation in the measurement system. Additionally, it is important to note that ASTM (D4435) highlights the precision required for such tests. Given the nature of the rock materials involved, it is often impractical or prohibitively expensive to produce multiple samples with uniform physical properties. Consequently, the focus is on minimizing variation stemming from sample properties, operator technique, or laboratory conditions (ASTM).
- 4.
The Evidence of an Inter-Laboratory Comparison:
The final requirement mandated by the accrediting body is an inter-laboratory comparison, as stipulated in the standard ISO/IEC 17025:2017 for precision assessment. Inter-laboratory comparison involves the measurement of the laboratories’ performance and the evaluation of test results of the same item or similar items by two or more laboratories under predefined conditions.
To fulfill this requirement, the company “Mining Industry X” was invited to participate in conducting tests on 5/8 anchors. The results obtained from these tests are presented in
Table 3. Subsequently, the repeatability and reproducibility calculations were performed, as shown in
Table 3.
3. Results
The results of applying the anchor test in mine A, with 3 operators and 20 tests each, are presented in
Table 3.
According to the Anova method used to conduct repeatability and reproducibility studies to assess measurement system quality control, if the % r&R is less than 10%, the measurement system is deemed acceptable for mine A. Additionally, this considers an uncertainty of ±2.2 tons and a coverage factor (k) of 2 for a confidence interval of 95.45% [
13].
The results of the anchor test conducted in mine B, involving 3 operators and 20 tests each, are presented in
Table 4.
Additionally, the Shapiro–Wilk normality test indicated a normal distribution for both mine A (with a value of 0.822) and mine B (with a value of 0.866). The results of the Kolmogorov–Smirnov test were not considered due to the small sample size (n < 50). It is worth noting that the Shapiro–Wilk test is generally more sensitive for detecting deviations from normality in small samples, while the Kolmogorov–Smirnov test is more suitable for larger samples.
According to the Anova method employed for conducting repeatability and reproducibility studies to assess measurement system quality control, if the % r&R is less than 10%, the measurement system is considered acceptable for mine B. This assessment is made considering an uncertainty of ±2.3 tons and a coverage factor (k) of 2 for a confidence interval of 95.45% (measured according to the guidance of the expression of uncertainty in measurement, ISO 1993) [
13] (refer to
Figure 5).
The results for the inter-laboratory experimental design comparison, developed in mine A with employees from ‘Mining Industry X’ who agreed to participate, are presented. The test involved the use of anchors of 5/8 with three operators, utilizing the same measuring instrument previously calibrated and used in mine A. These results are detailed in
Table 5 and
Figure 6.
According to the ANOVA method used to conduct the repeatability and reproducibility study for the quality control of a measurement system, if the percentage falls within the range of 10% ≤ r&R < 30%, the measurement system may be deemed acceptable for its intended use.
The smoothing densities, depicted in tons in
Figure 7, indicate that the variation between laboratories is insignificant. This statistical validation reinforces the presented method.
This demonstrates the reliability of the proposed method, as evidenced by the comparison of ANOVA tables, which show lower variation than that found in inter-laboratory testing. Regarding the margin of error, it is important to note that there is no accepted reference value for this test method, as per ASTM standards. Therefore, the margin of error cannot be determined. Consequently, the validation of the method relies on the acceptance criteria of the r&R studies.
The aforementioned results were submitted for accreditation audit in accordance with the current Quality Infrastructure Law in Mexico [
14].
4. Discussion and Conclusions
Documenting and validating a non-standard method can present one of the most challenging tasks for testing and calibration laboratories seeking ISO/IEC 17025:2017 [
1] accreditation. Many laboratories primarily work with standard methods, thereby focusing on verifying their test methods rather than validating them. Consequently, this process necessitates the development of a statistical knowledge base that laboratories may not have previously established, enabling them to achieve the required technical reliability.
The method chosen for validating the results holds paramount importance, carrying the highest level of credibility during accreditation audits, with ANOVA being the most reliable method at present.
However, the primary application benefits of the developed methodology are rooted in the Quality Infrastructure Law [
14]. Upon achieving accreditation for any method, laboratories transition into a Conformity Assessment Body. As defined, a Conformity Assessment Body is an organization accredited by an accreditation entity or, where applicable, by the Standardization Authority. In the case of Official Mexican Standards, international standards referenced therein, or other legal provisions, if accreditation is conducted by an accreditation entity, the body must be approved by the competent Standardization Authority to carry out the conformity assessment.
As stipulated in Article 54 of the aforementioned law [
14], only laboratories accredited by an accreditation entity are authorized to operate as Conformity Assessment Bodies. This provision establishes a national precedent in the geotechnical sub-branch of construction. The method presented in this paper stands as the sole accreditation as an OEC (Organismo Evaluador de la Conformidad) in Mexico at the time of publication. This information can be verified in the catalog of accredited members of the EMA, accessible at
https://www.ema.org.mx/portal_v3/ accessed on 28 November 2023.
It is important to emphasize that the reliability of results obtained by any other company using the proposed method will be subject to verification by the accreditation entity. Without such verification, the validation and reliability of the presented results will not be recognized, in accordance with the Quality Infrastructure Law [
14].
Future research in the field of rock engineering could delve into the implementation of ISO 17025 accreditation standards for testing laboratories engaged in assessing rock support systems. Investigating the impact of accreditation on the reliability and traceability of test results could yield valuable insights into enhancing quality assurance processes within rock engineering practices.
An area of potential research interest lies in the development of standardized testing protocols for evaluating the performance of rock bolts, cables, and dowels, adhering to ISO 17025 requirements. Establishing consistent testing procedures and measurement criteria can ensure the accuracy and comparability of test results across various laboratories. This effort ultimately aims to enhance the credibility and acceptance of rock support technologies within the industry.
Exploring the integration of quality management systems rooted in ISO 17025 principles into the design and execution of rock support projects presents a promising area for future research. Investigating the potential benefits of adopting a systematic approach to quality assurance and continuous improvement in rock engineering practices could lead to improved safety, efficiency, and alignment with international standards.
Furthermore, future research endeavors could concentrate on developing training programs and competency assessments tailored to personnel engaged in testing and installing rock support systems, in accordance with ISO 17025 accreditation standards. Enhancing the skills and knowledge of professionals within the rock engineering domain can ensure the proficiency and competency of the workforce, consequently leading to enhanced performance and adherence to quality standards in rock support projects.
Future research in the field of rock engineering could concentrate on the development of advanced grouting techniques tailored for rock bolts and dowels, aimed at enhancing their long-term performance in demanding geological environments. Exploring the interaction between various types of grouts and rock formations holds promise for devising more durable and reliable support systems for underground excavations.
Exploring the utilization of innovative materials, such as fiber-reinforced polymers, in rock bolts, cables, and dowels presents an opportunity to enhance the strength and durability of these support systems. Investigating the performance of these advanced materials across various rock types and environmental conditions could yield valuable insights for the future development of more resilient and sustainable rock engineering solutions.
Moreover, future research endeavors could concentrate on integrating monitoring technologies, such as sensors and remote sensing techniques, to evaluate the real-time performance of rock support systems in underground excavations. By implementing smart monitoring systems, engineers can gather valuable data on the behavior of rock bolts, cables, and dowels, enabling proactive maintenance and enhancing safety measures in rock engineering projects.