Getting from Data to Decisions that Drive Business Outcomes in the Mining Industry

Size: px
Start display at page:

Download "Getting from Data to Decisions that Drive Business Outcomes in the Mining Industry"

Transcription

1 Getting from Data to Decisions that Drive Business Outcomes in the Mining Industry 2018 DINGO DINGO.COM 2016

2 DINGO THE GLOBAL LEADER IN PREDICTIVE MAINTENANCE Delivering Results for Over 25 Years Human Expertise + Cutting Edge Technology THE GLOBAL LEADER IN PREDICTIVE MAINTENANCE World s Largest Fleet Performance Benchmarking Database $10B OF ASSETS MANAGED 236 OPERATIONS SUPPORTED Real World Technology-driven Workflows 130,000 + COMPONENTS CARED FOR $450M + DOLLARS SAVED Fast Implementation, Fast ROI Winner of the Mining Magazine Software Award RUN SMARTER 2

3 DROWNING IN A SEA OF DATA Less than 1% of available data in the mining industry is being used. Source: McKinsey Global Institute 3

4 INCREDIBLE UPSIDE TO GETTING IT RIGHT By unlocking the value of this dormant data, the mining industry could generate an economic impact of $370 billion per year in

5 OLD METHODS AREN T THE ANSWER A typical new haul truck is fitted with more than 200 sensors and generates 5 terabytes of data per month. Manual methods can t process this volume of data, while technology often creates more questions than answers. 5

6 BIG DATA IS POWERFUL BUT NOT PERFECT A major shortcoming of Big Data analytics is that it fails to explain why we do what we do. Tricia Wang 6

7 WHAT IS THICK DATA? WORDSPY - THICK DATA n. Data related to qualitative aspects of human experience and behavior, particularly when used as context for the analysis of a large data set. 7

8 THE IMPORTANCE OF THICK DATA What is measurable isn t the same as what is valuable. -Tricia Wang Volume Of Data More is better BIG DATA THICK DATA Maximizes on existing data; Placed in context Focus Technical cause / effect; Machine Business outcomes; People Decision Making Quantitative Quantitative + Qualitative Conclusions Patterns and insights from data Corrective actions to drive business outcomes 8

9 BRINGING IT ALL TOGETHER To form a complete picture, both Big and Thick Data are critical. The ideal system will capture, curate, and analyze both Big and Thick Data to produce outcome-focused insights and actions. 9

10 THICK DATA EXAMPLE IN MINING 10

11 GETTING TO OPTIMAL ACTIONS Detect - High Silicon, Iron and Aluminium - Probable dirt entry Big Data Answers Recommended Action - Clean, change oil and redeploy asset - Check with a resample Develop Insights - Typical causes of dirt entry - Type of machine, component, use case - Weather in region - One off event Human Insight 11

12 FAILURE PROBABILITY MODELING From the outputs of the anomaly detection algorithm, we created a predictive model to predict the probability of failure over time. It was tested against actual failure data to determine the appropriate confidence level of the model. Mill 2 Gearbox 2 Recommended threshold to catch failure 12

13 BIG DATA + THICK DATA Perform and document filter inspections on lube systems - Particularly when spikes in pressure differential are detected from sensors - Use filter magnets to get qualitative data Install, monitor and track results of magnetic plug inspections - Capture images of inspections to see the progression of issues - Use failure likelihood calculation to start triggering increased inspections at reduced intervals 13

14 TRANSFORMING INSIGHTS INTO REAL WORLD APPLICATIONS AND REAL RESULTS (*EXAMPLE OF RESULTS GENERATED FOR A DINGO CUSTOMER) Condition Monitoring Data Points Total Abnormal Observations Technology Rules Engine, Predictive Analytics, Machine Learning 410, ,867 Total Abnormal Reviews Total Reviews With Corrective Actions Total Breakdown Avoidance Events Human Subject Matter Expertise, Contextual/ Thick Data 102,356 26,074 1,820 REAL WORLD RESULTS $52,450,451 SAVED Total Breakdown Cost Avoidance $22,768,830 + Total Component Life Extension Savings $29,681,618 14

15 THE SOFT STUFF IS THE HARD STUFF To harness the power of their data, mining companies must break down internal barriers. The mines that are gaining a competitive edge are approaching data utilization as a business challenge not a technology challenge. 15

16 TRANSFORMING DATA INTO BUSINESS OUTCOMES 1. Start with the end in mind what are the desired business outcomes? 2. Bigger data and better technology aren t always better 3. Thick Data can help bridge the gap between Big Data and the real-world 4. Go together and focus on change management to achieve meaningful results 16

17 THANK YOU Paul Higgins CEO, DINGO For more information on how DINGO can help maximise on your data to increase availability, extend equipment life, and reduce operating costs, visit our website at or send us an at 2018 DINGO DINGO.COM 2016