/** Google Analytics start **/ /** Google Analytics end **/
Categories

The Impact of Digital on Oil and Gas Work

We all have an uneasy sense that jobs are changing rapidly, thanks to digital, but what will be the future of oil and gas work in an increasingly digital world?

Who cares?

After writing some 70+ articles about the impact that digital technologies are having (or about to have, or could have, or should have, or will eventually have, or might have), on the oil and gas industry, I’ve formed a point of view about the future of work as we know it. But a strongly held point of view isn’t as convincing as an actual study of the impacts. Fortunately, late last year, a clever chemical engineering student (Annie Nguyen – full disclosure – she joins Deloitte later this year!), contacted me to explore ideas on a potential research project that would be strong enough for academic credit and valuable to industry.

We landed on the Future of Work, which I was keen to explore in more depth after the first successful article on this subject, available here, and written by my good friend and former colleague, Dominika Warchol-Hann. This article has been read hundreds of times in the past 9 months.

Coincidentally, I had also been asked to participate on a panel discussion in March on the same topic, sponsored by the Schulich School of Engineering at the University of Calgary, along with several other interesting speakers (Bob Brennan, Jim Gibson, Andy Knight and Cherise Mendoza). It made sense to base my panel remarks on Annie’s paper, which I summarize in this blog post.

What work will be impacted?

We started with a hypothesis that industry would aim its digital innovation dollars at work that was some combination of extra dangerous (and so elaborate protections are needed to safeguard the humans), high volume routine, or extra costly (perhaps by virtue of its location or by the scarcity of skills required). To test the hypothesis, the study team interviewed 11 oil and gas industry professionals from around the world (more disclosure – they volunteered to my appeal for interview subjects on LinkedIn), and 7 university researchers in the field.

There was almost no difference in the likely impacts of digital on these three categories of work, but there was considerable difference in where digital innovation had made an impact already. High volume routine work has already received high attention and the deepest penetration.

There are a number of reasons behind the selection of routine work as the digital target. First, the more hazardous work or the work that was more costly to do was typically dependent on investing more capital, but the industry (in Canada at least) is still very capital constrained, making digital investment there less economic. Second, some work could be impacted by digital, but might be contingent on available infrastructure. Third, digitalising routine work tended to have the fastest path to value (ie, the returns could be realized within a year).

It turns out that there is low value routine work in virtually all industry sectors, which in turn means that no sector can rest comfortably under the delusion that digital is some passing fad to be ignored.

From drones in Amazon’s warehouses, to IBM blockchain on shipping containers, to Tesla’s self driving trucks – if a job involves holding a steering wheel, there’s a good bet it won’t exist in a few years. Public sector services, like business licensing, tax payer identity, self service business models will help reinvent the public administration of societies. New technologies like nano-robots, doctor-assist artificial intelligence, and 3D printed organs will help deliver improved health outcomes and extend lives.

While it was less clear when jobs in these sectors will be impacted, there was no doubt that the impacts would be dramatic and sudden. Transportation and logistics is likely first to feel change in a big way (in fact, it already is – two of Canada’s oil sands miners have announced automated truck projects).

The Human Hustle

If all work as we know it is going to be disrupted to one degree or another, then what is to become of the human worker? Put another way, what human attributes and work skills look like they will be least impacted by digital’s march into the work world?

Here’s a starter list of the human attributes that for the moment distinguish us from machines.

Leadership

The complexities of decision making might be assisted by machines, but humans look like they will own the leadership roles over other humans. That doesn’t mean all leaders are safe – the shift supervisor over haul truck drivers disappears when the haul trucks no longer have human drivers.

Empathy

Being able to connect emotionally with humans under a broad range of settings (think of the act of selling, providing coaching, delivering bad news, negotiating), looks like a restricted human skill.

Teamwork

Bringing lots of complex skills together in short bursts to work on complex problems collaboratively will remain with Homo sapiens. At the same time, traditional team work in factories is disappearing. Visit a shop floor – there are no humans about because the machines are clumsy and dangerous. The machines are often caged for our safety.

Communications

Basic communications are now being put together by digital solutions, but the ability to tell compelling stories is inherently human.

Customer insight

Machines are not yet able to draw deep conclusions about human motivations and personal drivers. Algorithms may be predictive (thank you, Cambridge Analytica), but if they were as good as humans, they would have predicted the backlash and considered the ethics of their analysis.

Innovation and Creativity

There are still many fields where humans are the supreme innovator, able to imagine new solutions to problems, invent new devices and create new art. Machines and algorithms have shown remarkable ability to improve on human design, but not yet to create the initial breakthrough.

Critical Thinking and Judgement

Digital systems may be able to pull together the data, and in increasing levels of detail and sophistication, to automate driving. But the decisions to start a business, to launch a product, and to close a factory, will always take humans to weigh the considerations and trade-offs.

My presentation was purposefully illustrated with hand drawings to make the point that in the future, work that involves story telling (drawing cartoons, telling jokes and weaving complex data and charts together), will not be displaced by digital. Sure, storytelling will be enhanced, but displaced? I think not.

Machines will learn. So must humans

One of the biggest digital innovations involves machine learning. At a simple level, machines learn through repetition. Group a thousand different pictures of forks into a category called ‘fork’, and it’s a good bet that a machine with an optical lens or scanner will be able to recognize a fork by quickly comparing an unfamiliar picture of a fork to the thousand other pictures of forks.

Machines are phenomenal learners. They have infinite capacity to learn, don’t take breaks and are programmed to stay on task for ever, or throughout their lives. Therefore they have what we might call high levels of self motivation. Machines are connected to one another, so that lessons learned can be instantly shared to other like machines. We connect them back to their makers to keep them technically current. We protect them with cyber firewalls and hide them in data centers.

For humans to maintain their place in the future of work, we must also adopt these same attributes. We must maintain our own curiousity and remain life long learners. We need to bolster our self-motivation to this task, because it’s in our self interest, and no one else’s. We need to maintain our connection to the learning institutions in our lives such as our universities, clubs and communities where learning takes place. We need to create the kinds of support we need for this task – sabbaticals, work tours, apprenticeships, job swaps – through the work world.

Conclusions

To summarize, I believe that all work is going to change, that key human skills and attributes will be permanently differentiating, but machines are made to learn, and so we must up our learning game.

You can find a copy of my presentation on Slideshare.net, but if your employer is one of those that blocks access to learning, you can just email me for a copy.

I owe a debt of gratitude to Jeff LaFrenz for proposing me to the organisers of the panel discussion, to Annie Nguyen for her stellar work on the paper, and to a dedicated team of researchers (Brooklynn Malec, Jenna Nguyen and Ibrahim Oshodi) for the interviews and analysis.

Share this ... Facebooktwittergoogle_plusredditlinkedinmail
1Comment

Leave a Reply

%d bloggers like this: