You are Home   »   News   »   View Article

EDITORIAL - Is the future of our industry model driven?

Tuesday, October 2, 2018

We are seeing much more use of digital models in the oil and gas industry. We don't usually think about 'models' as a subject in itself, but perhaps we should - if modelling is the way to get the best working relationship between people and machine - in all domains of upstream oil and gas - as well as the best approach to building software.

Continental plate models, geological models, reservoir models and simulators, drilling simulators and models, well models, production flow / decline models and simulators, 'digital twins' for topsides / equipment / maintenance, corrosion models, asset integrity models, 3D models, activity planning models, commercial models, personnel planning models, decommissioning planning models, graphs, are all examples of models in the oil and gas industry.

Model is an abstract term meaning a representation of reality. Models are also how people think and understand the world. We build models in our mind about how something works, how we are going to make decisions or behave when certain things happen. Another definition of an expert is someone who has well evolved mental models about their subject, so they can understand what is going on and make better decisions.

There's some big benefits to building digital technology around models. It presents information to us in a way which easily compliments the way we see it in our minds. It should be easier to evolve, combine and compare software models - just like the models in our mind. There is a growing technology called 'low code' where the code is generated automatically from a model - no programmers required. We are also seeing more and more companies offering 'platforms' on which different models can run on.

There could be many business opportunities building models, particularly for people who have domain expertise in a specific area and want to build a model encapsulating their expertise, which they can sell. For example if you have a model about how to adjust gas lift to multiple wells, or how to optimise the frac fluid you use, make the best decision about how to develop a new unmanned operation, or vet a new supplier. There could be markets where oil companies compare different models and choose the one which works the best for them, compensating the developer accordingly.

Modelling could make a big impact on safety management. Rather than be beholden to streams of data from sensors and your systems' ability to interpret it, a domain expert can build a model of what a safe operation should look like, and what an operation looks like when a problem is emerging but operations are not yet unstable. By comparing reality with this model can give operators early warning of emerging problems.

We use the term 'model driven' to talk about an approach to software development which starts by defining the model and then puts in the digital technology - rather than (how it is usually done) starting with the data, adding algorithms and then trying to build a model out of it. It follows an established discipline, 'model driven software engineering'.

We plan to explore the subject of modelling in more depth, in the Q4 issue of our magazine Petromall Insights, and also in our January 2019 London event.

If you have ideas or technology about how the oil and gas industry can do more with modelling, or are able to talk about modelling itself (not just talk about a model your company produces), please let me know.

Karl Jeffery, editor, DIgital Energy Journal

comments powered by Disqus


To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.


A modern data platform to improve predictions and company performance
Jane McConnell
from Teradata


Latest Edition June 2019
Jun 2019

Download latest and back issues


Learn more about supporting Digital Energy Journal