SHORT COURSES

Level: Introductory

Facilitators: Yuksel Asli Sari, Queen's University

This course presents an overview of the key elements of data science for engineers. The course will start by explaining the introductory data science principles and why it is needed in the mining industry. The big picture of data science, machine learning and how they fit into various mining applications will be elaborated to demonstrate the potential of these relatively newer approaches in current operations. The topics to be covered include preprocessing techniques, feature selection, exploratory data analysis and visualization, overall categories of machine learning techniques and how to choose an adequate approach given a dataset and a specific problem, and introduction to some useful, versatile machine learning approaches. There will also be an emphasis on the interpretation of the results. This will be followed by a demonstration of case studies where data science and machine learning principles are used on real datasets for prediction and analysis.

Short Course Objectives:

• Demonstrate the potential of data science and machine learning techniques for mining operational decisions

• Give an overview of the data science processes

• Inform the attendees on the necessary treatment of the data before the analysis.

• Introduce non-complex but versatile machine learning approaches that can be used by attendees later with their own data later.

• Present case studies to provide some example usages.

Target Audience:

Professionals interested in learning about data science and its application in the mining industry. The people who would benefit the most from this course would be mining, mineral processing and other engineers who work in mining operations, as well as managers who want to use their data for making important decisions.

About the instructors:

Yuksel, P.Eng. is an Assistant Professor at the Robert M. Buchan Department of Mining at Queen’s University in Kingston, Canada. She received BSc, MEng and PhD degrees from McGill University. With her background in computer science, she focuses on computer applications in the mining industry. She has developed tools for open pit mine planning (finding pit limits, block extraction sequencing and block routing) and stope optimization (stope layout planning, stope sequencing). Also, she has designed mathematical models for dig-limit optimization, open pit mine planning with landfilling option and to determine the stope limits. She has worked on developing machine learning approaches to dynamic haul truck dispatching, pillar stability in cave mining, predictive maintenance scheduling and decarbonizing grinding circuits. She has taught the courses “Applied Data Science”, “Applied Machine Learning” and “Introduction to Programming for Engineers” to students from mining engineering and other engineering backgrounds. Her research interests include short term and long-term underground and surface mine planning, data analytics and machine learning applications in mine optimization and mine automation.