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

Rethinking Oil and Gas Construction Using Digital

Digital innovation applied to construction could make the biggest impact on the economics of Canada’s LNG industry. Here’s how.

Setting up LNG to win

This past week I had the pleasure of hosting a couple of my Japanese colleagues in Calgary, who were visiting with a number of organisations to share how Japanese gas buyers view the LNG sector. This interested me because Canada totally missed the last window for launching any LNG projects, but the good news is that the window looks like it’s unexpectedly opening again. What Japan thinks is important because they’re the biggest buyer (some 80 million tons per year), and they set the market tempo.

Bluntly, Japanese gas buyers are not confident in Canada’s ability to build cost effectively new LNG export infrastructure. They benchmarked LNG cost structures across all of the planned export terminals (Russia, Qatar, Africa, Australia, Canada, American), and conclude that Canadian ambitions are still at the edge of the acceptable cost profile. In broad strokes, here’s the numbers they believe reflect Canada’s reality, all in dollars per million British thermal units (mmBtu), which is a measure of gas heat value:

  • Gas cost: $2 for BC gas (effectively the Alberta AECO gas hub price). This varies a bit, and has even gone negative recently, not that we want to give our gas away.
  • Transportation: $1 to get BC gas to the coast over a new pipeline (the 20 year toll to pay for the capital and operating cost). 
  • Manufacturing: $3-$4 to liquefy gas (a similar cost to the US Gulf coast toll, good for 20 years, includes power costs).
  • Shipping: $1-2 to ship LNG to Tokyo. This is set by the market for shipping and Canada can’t really impact this number.

This yields a cost of $7 to $9, for a 2 train, 16 million ton per year facility.

I could point out the dual Canadian advantages – the shipping cost from Canada as compared to the US or Australia is less (we are permanently closer to the markets, and we don’t have canal tolls), and Canada’s gas is lower cost (largely because the US has so much now that Canadian gas trades at a discount to US gas). We need to beat the American advantages – the US plants already have pipeline access, and they can repurpose existing regas plants to save a bit of capital.

The Australian LNG projects are similar to the US – adding additional facilities to an existing plant has a much lower cost because so much of the capital is reusable – jetties, tankage, lay down yards, work camps, utilities and off site assets.

Therefore, our big opportunity is in the controllable $3-$4 cost to liquefy gas. Can we do better, and if so, how?

One of my final projects before I left Australia in late 2016 was to benchmark the Australian LNG projects for competitiveness and to provide guidance on what Australia needed to do to improve its position. We concluded that an LNG project could cut its liquefaction cost by $1/mmBtu per year for a full 20 years by shortening the 4 year construction project to 3 years.

To put that in perspective, 1 million metric tons of LNG is 52 trillion Btu, so a 16 million ton plant delivered 12 months faster creates a pricing advantage worth $830m per year for 20 years. That’s plenty.

How can digital innovation shrink construction cycle time?

The builder’s standard playbook

The oil and gas industry has refined a number of tactical ways to manage the timelines for the construction of its facilities, starting with the Qatar developments in the 1990s, fine-tuned in Australia, and now coming to North America.

The engineering world is now highly digital, with designs completed almost entirely digitally. Instead of building the LNG facilities on site, they are built as a series of modules in overseas fabrication yards and are floated to the site, where they are lifted into place and welded together. Multiple yards can build multiple modules in parallel. Certain items with long lead times like steel are ordered early. Some expensive rental equipment (cranes, ships) that are in high demand get booked for specific times. Owners try to put in place union agreements that will last throughout project duration, and negotiate fixed price work packages with contractors. This pushes cost and schedule risks to contractors, but also can motivate contractors to remove from their bids anything that adds cost (like innovation).

My conclusion is that these tactics are solid contracting strategies, but digital innovation is not getting a look.

What about digital innovation?

To test my hypothesis, I met with two large engineering, procurement and construction (EPC) firms to ask about how they react to any digital innovation imposed by the owners. Their answer was “we raise our prices by at least 20%”. The reason is that the EPC industry has already invested in its own systems and methods for its purposes, which they are loath to strand.

While incumbent EPC firms are in a strong position to push digital innovation onto the construction sector, they don’t. As a result, some $1T of oil and gas infrastructure projects are stranded on the books.

Not surprisingly, the construction sector is frequently called out as being the least digitally enabled industry in the world. Construction productivity globally has actually declined by 20% over the past 2 decades, whereas most other industries have become more productive. Even the Canadian federal government doesn’t put much hope in the industry improving its performance – in the recent supercluster funding program, the construction industry lost out to fishing and farming. 

How Digital Helps Schedule

All the tools that could enable the construction industry to improve its productivity are available today, and all based on the same building blocks for all other industries (cloud computing, mobile devices, analytics, telecom networks, cheap sensors, robots). Here’s just a sample of how they could be configured to extract time out of the construction schedule.

Digitise the design

A cloud version of the plant design, and not just the workings of the methane chiller, but the whole plant, is an immensely important asset. It serves as the basis for quality and technical reviews, workforce mobilisation and stakeholder engagement. Reviews that surface design problems can eliminate rework. Orienting and mobilising the workforce faster cuts down on delays. Showing stakeholders what the plant will actually do removes resistance to the project. A digital design can be fed into an augmented reality engine to give users a far more visceral experience engaging with the design.

Digitise the plan

A great design without a great plan will likely take longer to execute. Industry leaders convert the construction plan to a kind of computerised video, which they subject to the same kind of quality and technical reviews, mobilisation and stakeholder sessions as the plant design. As much as 10% of the construction time can be cut out of the plan once engineers can more easily visualise how the plan works in practice.

There’s no reason why the digital plan could not incorporate every worker, every rental asset, every tool, every permit, every movement. This kind of big data problem has been crushed in many other industries.

AI the plan and design together

With the plan and design now linked up digitally, artificial intelligence can be put to work to analyse the plan and the design to figure out better ways to build and execute. The plan could be tested and run millions of times under any variety of specific goals and scenarios to see its behaviour and how it could be improved.

Build a digital version of the as built

A big issue in construction is that subtle design and build interface errors are caught far too late. An interface is where one fabricated asset has to line up with another asset. Industry leaders  fly drones over the modules in the fab yards and the corresponding interface at site to find those interface errors quickly. In this way, actual delivered assets match each other once they are mated, and delays are eliminated.

Inspect and supervise using Digital

Another issue in construction is that the scale of works can outstrip human capacity to understand and manage. Industry leaders are now using drones to fly over large construction sites to gather detailed data from the air, and build a digital version of work in progress that is used to inform the state of the build. Better data about current progress is vital to make better choices to keep a project to schedule.

Speed up movements and approvals

A big cause of execution delays involves the paper work required to move physical things across various jurisdictions. Customs forms, inspections, manifests, packing lists, etc. This is a perfect use case for blockchain, which can remove delays in moving items through a supply chain by making data common and shared.

Know where things are

Construction sites can be a bit chaotic, and finding necessary things that have gone walkabout is a constant problem. Not so in a digital world. Inexpensive sensors can be affixed to tools, equipment, assets and vehicles, to give real time visibility. It should now be unacceptable that we can’t see where construction equipment is on a site while we can see an Uber cab making its way to a pickup point.

Coordinate crews

Studies repeatedly show that field worker productivity in oil and gas is at best 50%, because of late starts, early quits, missing permits, missing tools, wrong crew skills, and so on. However, virtually every worker on a construction site now has a super computer in their pocket. These should be far better utilised to provide training, capture safety concerns, send and receive work instructions, issue permits, track tool usage, record hours, capture productivity data, and identify quality problems. Even the simple act of getting onto a site could be accelerated by using an app and a reader at the gate to let workers swipe in quickly.


Canada’s LNG projects could beat the Australians and the Americans at their own game, by accelerating capital project delivery via digital innovation.

Share this ... Facebooktwittergoogle_plusredditlinkedinmail
No Comments

Leave a Reply

%d bloggers like this: