COURSE TOPICS
Introduction
 Fundamental statistical concepts used in sampling theory and sampling practices
 Nine kinds of sampling errors: You must address them one at a time, otherwise sampling is almost always invalid.
 Heterogeneity of major constituents and trace constituents
 Examples of common financial losses due to poor sampling practices
 Definition of Data Quality Objectives
 Presentation of a new quality strategy based on Data Quality Objectives
 Synergy between Data Quality Objectives and sampling protocols
 Definition of basic terms and symbols
Sampling Theory and Practice

Errors
generated by sample weights
 Optimization of sampling protocols
 Description of Heterogeneity Tests, for a normal case, and for a difficult case
 Errors generated by segregation

Practical
implementation of sampling protocols
 Complete review of sources of sampling biases
 Exploration of the in situ Nugget Effect
 Selection of realistic, economical cutoff grades

Detailed
review of existing sampling systems:
 During exploration (Diamond core, RC, …)
 At mines (blastholes, …)
 At plants (cross stream systems, instream probes, augers, …)
 At laboratories (splitters, crushers, pulverizers, shovels, spoons, spatulas, …)
 For sampling commodities at shipping facilities
 For sampling the environment
 Monitoring precision and accuracy at the laboratory
 Monitoring precision and accuracy of sampling and subsampling protocols
 Quantifying the awesome cost of sampling precision
 Suggestions for better sampling standards
Reconciliation problems between the geological model, the mine, and the plant
 The myth of reconciliation
 Identification of major sources of reconciliation problems
 Capitalize on existing data: A gold mine of opportunities
 Understand the different kinds of heterogeneity and the variability they generate
 Become more proactive through effective statistical thinking
Management must set priorities
 Find causes of problems and structural properties you must live with
 Invest in minimizing causes of problems
 Find effects of problems and circumstantial properties you cannot control
 Save money by spending much less on effects of problems

Managing
visible cost:
 Historical priority placed on visible cost
 The accountant's point of view

Discovering
invisible cost:
 The staggering cost of constituents grade variability
 Reconciling statistical and accounting points of view
Introduction to Chronostatistics
 Critical review of sampling modes: random systematic, stratified random, and random
 Introduction to variography
 Advanced variography
 Introduction to variographic statistical process control
The Moving Average, a pragmatic, simple but delicate tool
 How much averaging is appropriate
 The random noise
 The corrected data
The Relative Difference Plot: The best tool for QC monitoring
 Detection of a conditional bias as a function of time
 Detection of a conditional bias as a function of increasing constituent content
An improvement strategy for effective sampling
Workshop practice using sampling software
Progressive workshops included in several lectures as the course progresses is the only way for attendees to take full advantage of what they learn so they can apply important principles as they return to their operations.
Participants will learn about:
 Installing the software packages
 Getting comfortable with the Help file of each software package
 Getting familiar with the many options in each software package
 How to prepare the necessary data from an Excel worksheet
 How to import the data to each software package
 How to customize the data analysis using each software package
 How to make a thorough interpretation of the results
 How to initiate or suggest new possibilities for sampling protocols