Imagine driving around a sprawling open-pit oil sands mine in a pickup truck, surrounded by haul trucks the size of three story houses. A truck coming towards you quickly makes a left turn down a dirt road towards one of the ore shovels, barely missing you. You later learn that these massive pieces of equipment are operating without a driver – part of an autonomous haulage system (AHS). The rise in development of autonomous, “self-driving” vehicles, as well as the high operational costs associated with mining, has led to the adoption of this technology in the mining industry .
In recent years, Canada’s oil sands mines have been the subject of environmental and societal controversy . On top of strict Canadian environmental and regulatory measures, the oil sands industry has been grappling with the worldwide drop in crude oil prices. As an operation with enormous upfront fixed costs, combined with high labor costs and operating costs, management constantly seeks to optimize for efficiencies. Automating mine equipment is a source of considerable savings with respect to labor costs and operational efficiencies [1, 3]. As surface deposits are typically located in remote areas and require above average labor costs, fly in/out operations, and residences set up for operators, automating machinery becomes an attractive alternative.
Suncor Energy, one of Alberta’s mining giants, has also made the case that AHS enables efficiencies in mining, while also improving safety and reducing collisions . As reported in the Canadian Mining Journal, “The greatest advantage of autonomous trucks is that they operate predictably, have prescribed route mapping, and employ obstacle detection systems. Reducing the interaction between people and equipment decreases accidents and potential injuries” . On January 30, 2018, Suncor announced that it will be staging a complete roll out of AHS over the next six years, resulting in the acquisition of 150 automated haul trucks that they believe will help the operation to run safely, effectively, and efficiently, but will result in the loss of around 400 jobs . Suncor also cited Australia as a successful example of deployment of AI in surface mining, benefiting from many of the above side effects of machine learning .
Machine Learning and AI in Oil Sands: Autonomous Haulage System (AHS)
Operating within the pressures of a low-cost environment as described above, management is turning towards AHS. Based on a simple GPS system, trucks need only know three things to operate: where it is, where it needs to go, and what to avoid . While Suncor hasn’t released which specific technology it will deploy, several patent descriptions available online describe the machine learning and continuous GPS monitoring of the truck along its path, with emphasis on collision avoidance and learnings to avoid “obstacles” through the application of buffer zones and safety envelopes of other mining equipment and light vehicles [7, 8].
With the current economic environment and future of oil supply in mind, my recommendations are:
- Suncor should continue to pursue its entry into automation of mining equipment past haul trucks and into other equipment, such as dozers and shovels.
- As an alternative to eliminating operators altogether from the operation, studies should be performed on the viability of equipment operators remotely operating from urban centers for more complex equipment (specifically, dozers and shovels).
- Eventually have all support staff (geologists and engineers) perform work remotely from urban centers, instead of on-site. This will further benefit the safety, cost efficiency, and mine productivity that is needed to stay competitive in this challenging environment.
Questions for Suncor and the mining industry at large:
In the near term, how will Suncor adjust the rest of its operations to be compatible with the new AHS technology? How will jobs lost be transitioned out of the organization and local economy? Will other technology upgrades be necessary to incorporate AHS into current mining operations and practices (ie, upgrading/changing current dispatch systems and collision avoidance technologies in light-duty vehicles)? While reduced operating costs will surely deliver better value for shareholders, how will Suncor respond to criticism for lost jobs in an already suffering industry and economy?
In the long term, how will Suncor maintain its competitive advantage and social license to operate as more aspects of its operations move to complete automation? As direct competitors in oil sands begin to adopt similar practices to Suncor, how will they hold their competitive advantage in this low-cost driven commodity space? As well, as Suncor and competitors continue to automate their mine equipment fleets, will governments allow them to continue to operate while depriving locals of employment opportunities?
Word count: 
 Nebot E.M. (2007) Surface Mining: Main Research Issues for Autonomous Operations. In: Thrun S., Brooks R., Durrant-Whyte H. (eds) Robotics Research. Springer Tracts in Advanced Robotics, vol 28. Springer, Berlin, Heidelberg.
 Charpentier, A.D., Bergerson, J.A., MacLean, H.L. (2009) Understanding the Canadian oil sands industry’s greenhouse gas emissions. Environ. Res. Lett. 4 014005.
 Kwame Awuah-Offei. (2016). Energy efficiency in mining: a review with emphasis on the role of operators in loading and hauling operations. Journal of Cleaner Production. Volume 117, pages 89-97. ISSN 0959-6526.
 Suncor’s website, https://www.suncor.com/en-CA/sustainability/innovative-technologies, accessed November 2018.
 “OIL SANDS: Suncor to phase in autonomous haul trucks over six years”. Canadian Mining Journal. January 2018. http://www.canadianminingjournal.com/news/oil-sands-suncor-phase-autonomous-haul-trucks-six-years/. Accessed November 2018.
 Bellamy, D., Pravica, L. (2011). Assessing the impact of driverless haul trucks in Australian surface mining. Resources Policy Journal. Volume 36, issue 2, pages 149-158.
 Burns, R.L., Parfenov, V. (2004). US6799100B2. Permission system for controlling interaction between autonomous vehicles in mining operation.
 Burns, R.L. (2002). US6393362B1. Dynamic safety envelope for autonomous-vehicle collision avoidance system.