Digital Mines ver 2.0 | 7 lessons on automation i learnt leading digital in a mining giant
0 likes76 views
The document outlines seven key lessons on mining automation and digital transformation at BHP, emphasizing the importance of data quality and human involvement in operational efficiency. It discusses the challenges posed by traditional organizational structures and advocates for strategic investment in people and technology to drive successful automation. The piece highlights the need for collaborative approaches and adaptive frameworks in navigating the complexities of automation in mining.
Digital Mines ver 2.0 | 7 lessons on automation i learnt leading digital in a mining giant
- 1. 7 Lessons on Mining Automation whilst driving Digital Transformation at BHP April 2021 with Coert du Plessis, DataAlchemists, +61406313111 @coertdup DataAlchemists Note: All views are my personal reflections and not endorsed by BHP or other organisations I work with
- 2. 7 Lessons on Mining Automation whilst driving Digital Transformation at BHP April 2021 with Coert du Plessis, DataAlchemists, +61406313111 @coertdup DataAlchemists ^ 6 Note: All views are my personal reflections and not endorsed by BHP or other organisations I work with
- 4. #1 data quality eats automation capital for BREAKFAST ∝
- 6. Data quality risk grows with each copy and store
- 7. Data quality and ownership is better in a network
- 8. Solving for Data Owners simplifies Data Quality 10x!
- 11. In a network of data, you’re a customer and a supplier Equipment Domain: 1. Data about comms towers 2. Data about fuel stations 3. Data about trucks Mine Planning Domain 1. Data about Mine Plan 2. Data about Truck schedules 3. Data about circuits Geospatial Domain 1. Data about stockpiles 2. Data about roads 3. Data about mine design
- 12. #2 beware how things are ∝
- 13. #2 beware how things are and fight CONWAY’s law ∝
- 14. Beware Conway's law Source: Conway’s Law https://www.thoughtworks.com/insights/articles/ demystifying-conways-law “Any organisation that designs a system will inevitably produce a design whose structure is a copy of the organization’s structure”
- 15. beware how things are… The expertise to schedule variable and highly adaptive people …does not translate to …a schedule for rigid and complex machines
- 17. #3 HUMANS are amazing! …and business cases contort to prove otherwise ∝
- 18. 2016 Paul Oakley, as we started up the innovation mine
- 19. Elon confirmed and owned it two years later
- 20. Business cases beyond the amazing humans • If your business case predominantly contains labor arbitrage - stop! You are in for a lot of pain – and risk • Uptime hours is obvious, but is the Mine Design, Planning and operating financial parameters also updated? Narrower roads? Driving times through blast risk zones reflected as shorter in the schedule? • Invisible costs? Networks uptime @99%. Missing the “other work” people ordinarily do. Sunk capital cost? Downtime from connectivity loss? • Who asks the hard questions? Peer pressure in the boardroom?
- 22. #4 people investment Gets you the MASTERY multiplier ∝
- 23. Mastery: Graham Reynolds Head of Autonomous Haulage Australia at BHP
- 24. People investment to Mastery = Investment Multiplier
- 25. #5 be clear how you play with others ∝
- 26. #5 be clear how you play with others who pays? innovates? learns? sell? ∝
- 27. The competition for automation is not in mining… Data from 2018
- 28. The OEM return on Investment in mining is not evident Of which 50,000 are big surface trucks (2020) Data from 2018
- 29. ~20 years = 1% of fleet automated in mining Of which 50,000 are big surface trucks (2020) Automated trucks ~800 Data from 2018
- 30. The Autonomy Market is fragmented in mining Urban Mining Investment $$$ à $5b+ VC $80b+ research $100+m VC <$1b research Leaders à Size of the prize à 1.4 b vehicles (UPS has 1.2m vehicles alone!) 88,000 mobile fleet 50,000 trucks
- 31. The “plays” are • Partnering with OEM all in • Partnering with part players (multi-modal) • Learning yourself, do yourself
- 32. #6 built to last ∝ SKIP
- 33. #6 built to last built to change ∝ SKIP
- 34. #7 everyone has to play ∝
- 35. #7 everyone has to play not to win… to know how ∝
- 36. Think different about the investment • Automation is inevitable -- $/t if done right means only some operators in the future • Approach: Copying “Digital Agile” verbatim can kill someone; you can’t be 80% right. Lean test and learn approach saves $. • Portfolio =/= equipment; automation portfolio = learnings • Cascading of automation knowhow: Trucks à LTE Networks à Dispatch automation à Fixed Plant automation à Mine Plan automation
- 37. Business case is investing in knowhow Today:
- 39. ∝ Questions