QA/QC

Quality Management (QM): The
heart of the QAQC process
Juan V. Salazar Jaime
Gerente Técnico Lab Perú Minerals
Juan.salazar@labperuminerals.org
-------------------------------------------------------------------------------------------------------------
MAY-2020 1
SALAZAR JAIME JUAN VICENTE
Ingeniero Químico
CIP N°158987
E-MAIL: jumavi1211@gmail.com
juan.salazarj@cip.org.pe
Ing. Químico con 16 años de experiencia en minería y otros rubros con posgrado en Ingeniería de la
calidad y Gestión Ambiental, con especialización en Sistemas de gestión ISO/IEC 17025, ISO 9001
,ISO 14001, ISO 45001 en la Pontificia Universidad Católica del Perú (PUCP) ,Universidad Agraria
la Molina (UNALM) y la Universidad Nacional Mayor de San Marcos (UNMSM). Experto Técnico
inscrito en el instituto Nacional de la Calidad (INACAL), Consultor, auditor, y capacitador con
sólidos conocimientos en la normativa vigente: Ley 29783, D.S.005- 2012-TR, R.M. N° 312-2011-
MINSA y sus modificatorias, D.S. 024-2016-EM y su modificatoria D.S 023-2017-EM, D.S 043-2007-
EM, G050, RM 111-2013-MEM/DM).
Con entrenamientos en Metrología Química en diferentes Institutos Nacionales de Metrología de la
región, como el CENAM de México, INMETRO de Brasil, INM de Colombia, INTI de Argentina, entre
otros. Ha participado como expositor en los Simposios de Metrología en el Perú.
Diseño , implementación ,ejecución y puesta en marcha de Laboratorio Químicos y metalúrgicos
Manejo y Gestión de costos y presupuestos ( SAP), dirección de personal, gestión y planificación de
materiales, control de Insumos y bienes fiscalizados (IQBF),mantenimiento y calibración de equipos
,Evaluación y manejo de datos estadísticos (Minitab, SPSS, Statgraphics), Implementador de
sistemas de gestión de información para laboratorios LIMS( Sample Manager, CCCLAS , Labware,
Global System ,Sapphire,Acme), Diseño, Implementación y puesta en marcha de laboratorios de
ensayos, experiencia de análisis en las diferentes técnicas como: AAS, ICP, IR, FIRE ASSAY,FRX ,
DRX, VOLUMETRIA, GRAV,ION SELECTIVO.
.

-------------------------------------------------------------------------------------------------------------
MAY-2020 2
Introduction
Business decisions are often driven by data, and for that reason data quality and reliability is paramount. In the
mining sector, investments in exploration, infrastructure construction, mining operations, ore processing,
transportation and port require multi-million dollar capital and operating budgets. The different phases of project
evolution are based on samples, usually a few grams that represent large tonnages. This was the observation that
led Pierre Gy to develop his Sampling Theory and later led researchers such as Dominique Bongarcon and Francis
Pitard, among others, to promote, convince, quantify and demonstrate to executives and mining professionals the
risks to which businesses expose themselves when compromising sample quality in a misguided attempt to reduce
costs.
In this context, quality assurance programs have been developed to establish Quality Assurance & Quality Control
(QAQC) parameters that monitor correct execution of sampling protocols and control each stage of the "sample
cycle ": sample collection, preparation (comminution) and analytical method. QAQC reports commonly include
statistical-numerical results that quantify performance of QAQC controls (field duplicates, preparation duplicates,
blanks, standards, etc.). Graphics such as scatter plots, QQ plots, histograms, and cumulative frequencies are
used to graphically represent the results. Statistical values including relative difference, absolute difference,
relative variance, averages, AMPD, T-test, and Z-scores are used to quantitatively express the relationship
between duplicate pairs... however, is an effective quality program just a statistical exercise? The following
discussion considers this question in the context of a quality program standard as outlined by the JORC code,
trying to highlight the call to return to the basics during this era of new technological applications and advanced
statistical analysis.
The case for proactivity
This paper aims to highlight the concept of "Quality Management" (QM) as the precursor of corrective actions
closing gaps determined by trend analysis (by ranges time and/or grades) with the aim of proactively determining
control performance deviation and thus proactively rectify the source of deviation.
There is sometimes confusion among those accountable for quality assurance, and even among auditors, that if
individual data points fall within a predetermined acceptance limit then they are necessarily acceptable and
therefore suitable for informing operational and investment decisions. A similar situation is that tabular summary of
statistics is enough to demonstrate acceptability of quality control outcomes. However, what is stated with respect
to QM is that sometimes results found within the acceptance limits can indeed be internally biased, or show
material deviations over a period of time, thereby impacting operational performance. An unstable process which
happens to plot within arbitrary acceptance limits is nevertheless an unstable process. Thus true process control
requires something more.

-------------------------------------------------------------------------------------------------------------
MAY-2020 3
QM refers to the proactive detection of these "anomalous tendencies"; that is, the trend over time/grade of a given
statistic. QM includes also the process by which these trends are understood, communicated and rectified. Some
businesses refer to this process as “continuous improvement” or the “Plan, Do, Check, Act” cycle.
This proactive approach in the mining industry can have material impact on financial outcomes through sequence
optimization, contract negotiation, and management of plant and processing infrastructure.
Below are examples of how QM can be implemented through the mining value chain, using a proactive approach
as guided by JORC Table 1, and how results are typically presented in QAQC report or audits.
1) Sample Collection
JORC Table 1 provides guidance that drilling campaigns shall deploy measures to maximize sample recovery and
representivity. A typical example for an RC drilling campaign would be actual sample weights measured against a
theoretical “ideal” drilling recovery, as a function of material density, rod length and diameter, and aperture size of
the sample shoot. Where duplicate samples are collected, it is expected that they will have similar, if not identical
sample weights. This is an indication that the rig set-up, sampling devices and drilling/sample collection process
are operating according to design.
Results are commonly presented as in Figure 1, where a scatter plot shows the distribution of the results between
duplicates. In this example, the scatter plot shows differences in weight outside expected thresholds, between 10 to
30kg; and potentially a small bias towards to sample A being heavier than sample B..
Figure 1. RC Field Duplicates performance: (A) Scatter plot comparing duplicates sample weight.

-------------------------------------------------------------------------------------------------------------
MAY-2020 4
There are several questions this graph fails to answer: Why are A samples systematically larger than B samples?
Is this the consequence of a particular drill rig? A particular sampling device? When was the bias first introduced?
Is the bias random or sustained for a period of time? What was done to fix it?
Figure 2, presents an example on how QM practices can proactively improve sample collection by monitoring rig
performance in a different way, whilst still comparing the weight of duplicate samples:
This graph can be interpreted as follows: During the first 2 weeks of drilling in February, weight differences in rig 1
were not performing within accepted thresholds (Relative Difference + 20%). A conversation with the drill crew and
drilling company supervisor is conducted in the field to explain to the driller the importance of drilling on geological
models, to understand the sources of this poor performance, develop an action plan to improve the sample
collection process and obtain their commitment to increase the quality of the samples.
Figure 2. Example of monitoring sample weight on duplicate samples. Quality Assurance (QA): Collect
sample weight on Duplicate samples. Quality Control (QC): Sample weight within +20% relative difference.
Quality Management (QM): Continuous monitoring of the information and actions were results are outside
expected thresholds.
Through QM, corrective actions are taken by continuously monitoring results over time. This proactive approach
can save thousands of dollars by “doing things right the first time” rather than review QAQC performance en masse
once the drilling campaign is already finished, by which time it is too late!!!
2) Sample Preparation

-------------------------------------------------------------------------------------------------------------
MAY-2020 5
Following the same criteria as Sample Collection, the JORC Table 1 benchmark requires evidence that “quality
control procedures [are] adopted for all sub-sampling stages to maximize representivity of samples”.
Usually, Blanks and Duplicate samples and sizing tests are used as a QA tool to monitor the performance of
crushers and mills. Later, results are included on QAQC reports where the performance of crushers and mills are
summarized (for example) as shown in Figure 3.
Figure 3. Examples of how Duplicate sample performance is presented in QAQC reports.
While these graphs and summary tables are typical, this information doesn’t allow us to apply Quality Management
(QM) to monitor the information in real time and proactively improve the results.
Figure 4, shows an example where a trend analysis is performed on a time (date) and grade basis: A) The Absolute
Difference of Duplicate samples is plotted against the date the Laboratory has reported the results. The graph
doesn’t show major issues over a specific period of time, but if the data is assessed on a Grade basis as shown on
B), a trend can be interpreted as the grade of the primary sample being greater than the duplicate sample. The
action here will be to talk to the drilling company if these are field duplicates; with the team performing the core
cutting, or with the laboratory if they are crusher or pulp duplicates, to find the source of this bias, and develop an
action plan to fix and close the gap. This real-time assessment and management is the basis for a proactive
approach. It needs to be highlighted, supplementing reactive activities such as reconciliation results or
monthly/quarterly QAQC reports (if done), where the opportunity for fixing issues in near-real time is lost.

-------------------------------------------------------------------------------------------------------------
MAY-2020 6
Figure 4. Examples of trend analysis performed on a Time and Grade basis for Duplicate samples
(applicable for field, crusher and pulp duplicates). These graphs highlight the value of performing QM on a
date and grade basis: the analysis by date doesn’t show any major issue in terms of bias and the results
look consistent, but the analysis performed on a Grade basis highlight a bias at high grades that needs to
be reviewed, understood and fixed.
3) Chemical Determination
Certified Reference Materials (CRM) are used to monitor laboratory performance, where mining companies shall
arrange preparation of their own CRMs to perform QM. It is not recommended to rely on Lab internal QAQC
processes. Changes in the lab results or consistent biases across time are best detected by an internal team
accountable for QM, in order to highlight issues with the lab, analyze the sources of deviation and consequences to
production, generate an action plan and apply lessons learned to avoid repetitive issues.
Usually statistical analysis considers “average values”, which sometimes lead to inaccurate conclusions that
assume a process is “on average” controlled or “fit for purpose”. Quality Management applies a different approach,
assessing data in real time, thereby escaping the need for averages, and keeping a business focus with the aim to
ensure consistent results supporting sustainable business decisions.
Figure 4 demonstrates the differences between an approach reliant on averages and Quality Management applied
to CRM results (QA= CRMs, QC= + 3 SD & QM= Trend analysis). Figure 4A shows 10 months performance of a
CRM. Because results have been performing mostly within 3 standard deviations, the business might infer the
process is well controlled and feel confident given the global average is close to the certified value. However,
Figure 4B shows the internal variability which the laboratory (period average) is observing over time. This lack of
consistency gives rise to operational instability, exposing the business to risks of under or over performing at
production, processing and compliance to plan results, or variable products
Such are cases where Quality Management becomes important by monitoring information in real time and
detecting changes in the performance of the Laboratory proactively, thereby ensuring consistency and
sustainability of business results.

-------------------------------------------------------------------------------------------------------------
MAY-2020 7
Figure 4. Certified Reference Material performance showing results performing mostly within 3 expected
standard deviations. A) Global average is very close to the certified value which can be interpreted results
are considered valid. B) Period average has been included, showing the big variability on lab performance
during months.
Conclusion
This paper aims to highlight that a Quality Program is not just a statistical exercise, where global averages or
standard deviations assure sustainable and consistent QAQC results. Examples provided in this paper
demonstrate the value of Quality Management to complement routine QAQC processes and statistical analysis,
where a proactive approach and data monitoring can really ensure consistent results across time or a range of
grades, and reduce resource and operational risks.
Indirectly this paper highlights the value and necessity of having a centralized team accountable for governance
and performing quality-related activities (QAQC & QM) across Exploration and Production.
Finally, it is in vogue and companies has been pushing to be part of a new era of new technological applications
(sensors) and data analysis (machine learning, conditional simulations, etc.) trying to provide businesses real time

-------------------------------------------------------------------------------------------------------------
MAY-2020 8
data to be used for business decisions in real time.….This paper highlights that either new technology or advanced
statistical techniques, need to be based on good quality data to calibrate and test tool’s calibrations, and quality
data needs to be incorporated into simulations or advanced statistical tools.…Quality Management becomes more
relevant to ensure that performance of future technologies will are robust...

QA/QC

Recommended

Recommended

More Related Content

What's hot

What's hot (14)

Similar to QA/QC

Similar to QA/QC (20)

Recently uploaded

Recently uploaded (20)

QA/QC