Petrobras unveiled its business and management plan for 2018-2022 in December 2017 that includes a chapter on capturing the opportunities offered from digital transformation across the value chain, from geoscience, reservoir management, automation, artificial intelligence, big data and the cloud. Oil IT Journal interviews Petrobras’ Nelson Silva (chief strategy officer) and Augusto Borella (general manager for digital transformation).
Oil IT Journal - What role does digital transformation play in your strategy?
Silva – Our 2018-2022 business and management plan introduced three new strategy target areas, preparing for a low carbon future, an increased focus on the digital transformation and evaluating and mastering financial risk.
Oil IT Journal - Lets focus on N° 2, the digital transformation.
Silva – Sure. We currently have a lot of initiatives across the company from upstream to downstream. These are coordinated from a central unit, the digital transformation area with the objective of minimizing bureaucracy and increasing speed of development. A huge effort has gone into understanding what is going on across the company, mapping many independent initiatives to form a clear picture of what we already have. We also have looked elsewhere, at what other oil and gas companies are doing and what other industries are up to in the field. We are now in the final stages of collecting these lessons-learned and are developing our final proposal to the board. Meanwhile, the digital transformation is ongoing, as we consolidate our activity, avoiding competing initiatives and conflicting efforts.
Oil IT Journal - Can you give some examples?
Silva –We are targeting the ‘moonshot.’ We are seeking disruption! Usually, folks do not have the time for this kind of thing. We are not only targeting incremental change although that will continue. We are looking at agile teams and methods as well, to reduce drastically the time-lapse between exploration and first oil. We are not talking about down to six months or a year – we need to be able to compete with US shale.
Oil IT Journal - OK – but that goes way beyond digital.
Silva –Yes it does, but the digital transformation is really about doing business differently and it does go beyond just ‘digital’ transformation. It is about using all available technology differently. People at the coal face don’t have time to do this kind of thing.
Oil IT Journal - OK, so how are you going about it?
Silva –A good question. But that is why we call it a moonshot! We will be giving targets and challenges and providing an environment where our engineers can try things and where they will be free to fail too. We live in a risk-adverse industry, so we need to test things that are not usually tried. Think back a couple of decades, before the iPhone and other modern technology marvels, and you can see that we all now behave completely differently.
One specific idea addresses the huge resources we devote in complying with government agencies’ queries. This currently involves a big head count. Why can’t we replace these with intelligent robotics that exchange information with robots at the agencies?
Oil IT Journal - You mean machine-to-machine reporting?
Silva –Yes. We are prototyping continuous machine-to-machine communications and discussing this with the federal government which is very receptive. As you know, recent Brazilian history has been plagued with corruption issues and we see this as a means of avoiding such problems in the future.
Another field is seismic imaging. This currently suffers from an excessively long workflow. It can take a year to process and interpret a large survey. We want to shorten this down to minutes! There are many other similar opportunities for similar ‘moonshots’ in oil and gas. Of course they might not all work! No Limits!
Oil IT Journal - We have reported on a few initiatives to use machine learning to go straight from raw seismic data to a reservoir model. Is this the kind of thing you are looking at?
Silva – Exactly. We are under pressure to compete and shorten the time to market. Data science is very important. And it also has potential application in the field of safety, in maintenance and in predicting equipment failure before it occurs. Another use case is knowing where people are and controlling access and operational behavior. AI and analytics can help with asset integrity, doing stuff on our behalf.
Oil IT Journal - Any noteworthy conventional earlier projects to report?
Borella – Yes. For instance, in our R&D center we have developed an image recognition/processing system that can identify and track people in video imagery. We also work with drones for data acquisition and imagery – including underwater. We also developed a state-of-the-art control room for real time drilling operations which has a machine learning layer on top. One example is our PWDA (pressure while drilling analysis), a real-time data visualization platform that was developed to support drilling but that can be applied in other areas. The platform monitors real time pressure data from sensors on the rig and an artificial intelligence algorithm receives and interprets the data. This warns of possible operational problems and drilling performance issues. Since mid-2014, Petrobras has tracked the results from the program showing almost $100 million of losses have been avoided with the use of the platform.
Oil IT Journal - Who is doing this work – do you work with Brazilian R&D orgs?
Silva – We have 1,500 engineers in our own R&D center, our primary source of know how and brain power. But yes, we are in the process of selecting out technology partners. We will have partnerships with technology firms and consultants along with our own resources. We are also seeking expertise in-house. Our early internal announcement on the new digital structure brought a great response, different from the usual structure announcement. Lots of young people in the company said, ‘we want to be part of this!’
© Oil IT Journal - all rights reserved.