Mars Khasanov, CEO of Gazprom Neft’s Science and Technology Centre, talks to Stiumul about maintaining industry-wide technological leadership, what effective corporate R&D actually is, and why you don’t need innovation for innovation’s sake.
Mars Khasanov is one of Russia’s most authoritative experts in oil production, Head of Gazprom Neft’s Technology Directorate since 2011 and heading up the company’s Science and Technology Centre (STC), which celebrated 11 years in operation on November 15. It is largely down to his efforts that market players cite Gazprom Neft as Russia’s most technologically advanced oil production company.
As well as being a successful senior manager, Mars Khasanov also has extensive experience in R&D: the key areas of his research being modelling, the control and management of non-linear multiphase flow, and developing computer technologies for design and process monitoring in oil field development. He was responsible for revealing the principles behind the transition to chaos in non-Newtonian (quasi-viscous) movement, and space—time fractality in oil production processes. The engineering methodologies and software developed under his leadership have been used in the design and development of some of the most important fields in the country, including the Priobskoye and Prirazlomnoye fields, and the Messoyakha and Novy Port projects — one of the most important new fields in Russia commissioned in recent years.
— What were the factors behind Gazprom Neft adopting its Technology Strategy in 2014?
— The main factor was business need. All of those reserves discovered and developed in Soviet times are becoming increasingly depleted. And the time is coming — for our company, as much as for the oil industry as a whole — at which we need to develop new kinds of oil and gas reserves, as well as new regions. So we have several challenges we need to address. The first being — how to increase oil recovery (the oil recovery factor, or ORF) at already drilled-out blocks. The second — developing new blocks and regions, which is problematic in as much as the filtration loss here is hundreds of times worse than that of traditional fields. We can only develop these hard-to-recover reserves profitably through new approaches and technologies. Added to which, when we talk about new technologies, this doesn’t just mean new materials and equipment, but new workplace management strategies and design engineering.
In the process of developing our Technology Strategy we identified a range of potentially industry-disruptive issues. Which is how we formulated the nine strategic areas. So, based on ongoing challenges, and those we saw on the horizon in the next decade, we decided, in 2014, to update our approach to technological development, and incorporate it as part of that strategy. It’s pretty ambitious —by 2025 we expect to bring more than 100 million tonnes of additional reserves into production, with cost savings of more than RUB100 billion.
— Four years on from the adoption of the Technology Strategy, you’re in a position to draw some initial conclusions. What are they?
— As well as priority objectives, our Technology Strategy included creating a system for managing the company’s technological development per se. In which context, our ultimate objective is the ongoing development of a culture of innovation — to make it possible for every company employee to be an innovator, an inventor, a creative individual: to have a system that scooped up any good ideas, combined these with others and, on that basis, created innovative approaches and technologies.
Added to which we have already, in a relatively short time, been able to put in place integrated development programmes across all nine areas — which are, essentially, mini-strategies. On that basis we’ve launched about 150 technological projects, some of which have, already, delivered tangible results — most notably in the construction of high-technology wells with extensive horizontal sections.
— Do you plan to make any amendments to your Technology Strategy?
— We’re updating it now, to what will be our “Technology Strategy 2.0”. We don’t plan any major changes, we just want a greater focus on multifunctional projects. These need, concurrently, to take on board the issues of new materials and digital technologies, in the light of optimising oil production and enhancing oil recovery, exploration, and effective well construction. This is all directed at securing a tangible outcome in terms of improving profitability in field development.
— You say you want every employee to be creative — isn’t that scary? Engineers’ experiments aren’t always successful
— I’m talking about creative ideas, not creative activities. Any idea has to be tested. We have an entire system for selecting and evaluating ideas, whereby they are assessed by specialists. If an idea seems interesting, that person is given a team or financing to implement it.
— And what does the inventor get?
— A bonus — albeit a small one — but, more importantly, recognition and an opportunity for personal fulfilment, a chance to get a bit of a boost, socially.
— The projects you’re talking about — are these being implemented mainly within the STC, or are they being outsourced?
— We want to work on the principle of “you pull the trigger and you ride the bullet”: you pull the trigger with a tiny touch of your finger, but get a mighty blast and a bullet shooting some way off. We understand that all of our tasks — the number of which is only going to grow — can’t be addressed independently: we need to engage Russia’s — and the world’s — full R&D potential. The success of our Technology Strategy will be determined by whether or not we are able to involve a large number of innovators to work on our tasks, and whether we’re able to cooperate effectively with them.
Just stating the problem isn’t enough. Outsourced engineers, left to their own devices, are unlikely to be able to deliver a finished product. Anything useful is only likely to come from close collaboration. We need to work closely with them. And at the initial stages. Our involvement needs to be highly proactive in setting tasks and outlining objectives: setting up and managing projects like this — it’s an art.
— Who are your main partners in joint projects?
— Service companies and higher educational establishments. Russian ones include, for example, Moscow Lomonosov State University, Lomonosov Moscow State University, the Moscow Institute of Physics and Technology, St Petersburg State University, St Petersburg Polytechnic University and Tyumen State University — all of which have very good backgrounds in mathematics, physics, technology and engineering. Our collaboration with these is pretty extensive. Together with various academic centres we are implementing about 80 R&D projects, and investing about RUB350 million, every year. And about 100 students complete internships or secondments at our STC every year.
— Do you do any work with start-ups?
— Of course. We want to get make contact with every centre, and large and small innovation teams, working on areas of interest to us, and understand their skills and competencies. And these areas include not just traditional issues in the oil industry, but also areas not directly related to it: new physical effects and fields, innovative materials, and new techniques in mathematical modelling. We are thinking about how to involve unconventional and breakthrough things emerging in other areas, and how to nudge others towards this.
— I’ve had to deal with start-ups a lot. Their main complaint as regards the oil industry is that the industry isn’t very enthusiastic about innovations, and that while you can often implement innovations abroad, you can’t get through to domestic oil producers. What should start-ups that want to work with you do?
— There are a lot of myths regarding the oil industry. I’m not aware of any genuinely useful start-ups that we don’t listen to. Anything that is genuinely interesting and workable we’ll always engage with. And with all due respect to innovation companies, their technologies are never completely ready — very often they’re just ideas. Start-ups, physicists, and mathematicians think they’ve put forward a brilliant idea, but we understand the limitations that mean it’s not going to fly. Ideas and technologies need to be fine-tuned and firmed up, and that’s often a long and difficult job. When you tell them that, people turn round and leave. It is important for start-ups to be open and ready to listen to business partners, and to adapt their product quickly and creatively in conjunction with them.
Added to which, we get shown a lot of pseudo-discoveries — including, for example, explosive-based EOR techniques. They tell us “Fracking’s expensive, and needs replacing — let’s use explosives.” But the oil industry is extremely sensitive when it comes to industrial safety, and setting up explosions within wells really isn’t the best idea. Our company’s advantage lies in the fact that we have many physicists working with us who have a clear understanding of these processes, the forces that impact them, and the effects they produce. They can often tell straight away whether an idea’s going to work or not.
— What made Gazprom Neft decide to set up its own STC, and not rely on third-party service companies?
— We at the STC are doing exactly what all of the major global oil companies are. Work like creating geological models and conceptual developments is always undertaken by oil companies’ employees themselves, this is never outsourced. About 70 percent of our work relates to those kinds of activities. In the same way, all oil companies independently develop their own technological development strategy, commission R&D and engineering, and take on board the outcomes of these. In order to become a client — in order to commission these things — you need your own internal specialists: it isn’t possible for one individual to work with a hundred institutions.
Besides, all oil producers have in-house research laboratories working on things that may become know-how and secure their competitiveness. If you outsource this function to third-party designers, they’ll complete the job and rush to sell it to somebody else, and you’ll lose your competitive edge. Priority areas here include digital solutions for formulating and taking investment decisions, and for managing investment project portfolios. We develop these solutions on our own because they represent our know-how and determine our company’s efficiency and competitive advantage.
Also, the STC runs a Drilling Control Centre that operates round-the-clock, in four shifts: two of which are tasked with continuously updating geological models of wells, and a further two of which are responsible for updating engineering calculations and designs. The Drilling Control Centre supported the construction of over 850 well in 2018 — which means all of the high-tech wells drilled by our company this year. We have about 1,000 employees working at the STC today.
The optimum needs to be global
— You’ve cited economics and mathematicians among the specialists working at the STC, but at what point does economics come into play in selecting a specific technology?
— We bear economics in mind at every stage. A major mistake on the part of several Russian oil companies is that they initially develop some technological scheme — for example, developing some field or another — and only then consider its economic efficiency. This is the wrong way, because in developing any technological strategy you’re always optimising. You have to select the optimum length for the horizontal section, for well spacing, for drilling speeds, and for the through-put capacity of surface infrastructure.
From the economic standpoint, that optimum has to be global: that is, it has to cover the entire system, overall. If you make an initial choice, based on experience and intuition, and only then start thinking about money, then you might be able to determine whether that project is likely to be unprofitable. But you’ll never have any guarantee that you’ll get the greatest possible profit from that project. Which is why economics is always at the forefront in all our calculations, right from the start of working on a field, and we look at its development in its entirety, taking into account not just drilling and direct production, but all infrastructure costs — which have a strong impact on choosing specific solutions.
Previously, it wasn’t that important: the size of fields meant infrastructure costs weren’t that important a factor in the budget. But now, with the move towards smaller fields, which are half the size (or even smaller) than those in middle of the last century, the situation has changed radically. You can, of course, optimise things once you go into operation: but that’s only going to improve efficiency by 10 to 15 percent. Whereas, at the concept stage, you can improve things by 50 to 90 percent, exponentially reducing costs or increasing production. And yes, we too have, ultimately, begun to calculate efficiency more accurately.
— Which of your current projects do you consider to be the most impressive and successful?
— More than anything else, our success in building high-technology wells — horizontal wells, including with multiple offshoots — and multi-stage fracking. You need a whole range of design, support and geo-navigation solutions for this.
The second thing we’re most proud of is the alkaline-surfactant-polymer (ASP) flooding project, implemented in conjunction with Salym Petroleum Development. Surfactants are surface-active compounds that produce a soap-like foam. Surfactant-based mixtures are injected into strata at mature fields to enhance oil recovery. Many people had little confidence in this technology because surfactants didn’t really prove themselves under the Soviet era. There was a lot of noise around surfactants initially, and two plants to produce these compounds were almost built in Bashkortostan, but it ultimately came to nothing because these surfactants were found to leave deposits on the surface of porous media. So we now have an entirely new generation of surfactants, and, with our partners, have synthesised 11 new formulations, and undertaken field experiments. We’ve been able to show that oil recovery (the ORF) can be increased by 17 percent. For example if, under normal conditions, we can extract 50 percent of the oil from strata, then using a new composition, it’s 70 percent — an enormous total increase in absolute terms. The impact of using this technology throughout Gazprom Neft could be in the order of 250 million tonnes’ additional production.
Another area we’re particularly proud of is our projects in digitisation. We’d already initiated these in 2012 — some time before “digitalisation” and “digitization” became buzzwords. We named this our “Electronic Asset Development” programme" or “ERA” (reflecting the Russian “электронная разработка активов” — ed.), and it’s extremely important to us. Oil companies don’t build wells themselves — our contractors do that. Our main concern is taking the best possible investment decisions. And for that you have to work with enormous volumes of data — the STC alone generates about half a terabyte of data every day — meticulously analysing this and building precise models. To that end, at the exploration stage, we also build a “digital twin” of a field and then work with it at every stage, from conceptual development to oil or gas production, fine-tuning and improving the model.
Over the last few years we’ve created an entire ecosystem of proprietary software products, operating within a single and cohesive information space. We started with data, and now have a full digitally remastered database that’s very easy to work with, with integrated management tools. We then set up automated (computerised) workstations for our geologists and design engineers developing reservoir management techniques, and for concept engineers developing field management strategies. Fifty percent of these involve unique working methods, a completely unique code, an ultra-user-friendly interface, and a direct link to our data base. We’re now in the process of developing the entire ecosystem. For example, our “Cognitive Geologist” project envisages the creation of self-learning models of geological formations. This saves time in analysing data, down from two years to a matter of months in analysing thousands of options for field development and selecting the best one.
Ideally we’d like to develop a single, integrated platform, the cornerstone of which will be our corporate database, to which our own and third-party solutions can be “built in”. We’re already moving towards this. Various service companies offer these kinds of offer platform solutions, but using them means they’ve got you on the hook for the rest of your life. We don’t want to be dependent on anyone. Added to which, these solutions often come from the west — which means they come with potential risks.
— We’re talking a lot about your achievements: but have you had any failures?
— We don’t make unrealistic plans. So I can’t say one area or another has been wrongly selected and has led to a major setback.
But a trial-and-error stage is typical in any R&D. There are no guarantees in any business, particularly not in those sectors involved in inventing something new. For example, we began developing a new fracking technology with one of the innovation companies. It was interesting and held considerable promise, and we gained a lot of experience from it. There were a lot of setbacks, initially, but then we began seeing some success, and made it replicable. But we weren’t able to deliver extensive replicability because the technology turned out to be too expensive: the production cost of the oil extracted using this technology was more than the cost of using traditional techniques. It was made clear to us, once again, that in developing something from scratch you have to think about rolling it out, and have to determine the necessary economic indicators right at the very beginning.
— Are there any specialities whose functions could be completely replaced by automated systems?
— I’m particularly sceptical of that, simply because it’s much cheaper to do without it. Automation isn’t an end in itself — something people are forgetting that in all the hype around digitisation. We’re not interested in digitisation so much as improving workplace efficiency and production. Are we likely to make serious economies by fully automating drilling and getting rid of the supervisor responsible for making all the decisions? No. Compared to other expenses in oil production, manual labour costs almost nothing. Work involving risk though — firefighting, for example — is something else entirely. In that situation it’s worth using robots instead of people. Manual labour represents only a small cost: but human life — that’s priceless.
When it comes to automation, you have to find the right balance: and I think this lies in putting in place effective man—machine dialogue systems whereby people make the decisions, and the machines implement them — doing all the rough work on calculations, information retrieval and linking data streams.
Seeing the unseeable
— You place considerable focus on modelling in oil production. What are the main challenges here?
— The main problem is that we can’t see what we’re modelling. A field is an enormous thing, running to dozens of metres, while the thickness of individual layers (which typically number about 50), is small and, these days, likely to be just a few metres. In some places you’ll find water, in others — oil, and in others -gas. And all of this at a depth of several kilometres.
In the exploration phase we take seismic profiles and drill 10 exploratory (prospecting) wells. Exploration wells comprise 20-cm-diametre boreholes, spaced about 10 kilometres from each other. They give an indication of what lies below the surface, but by no means give the full picture. And in seismic surveying we produce sound waves which are refracted, echoing and bouncing back from numerous strata. In order to produce an accurate picture, you have to determine what each wave is being deflected from: and that involves an enormous amount of calculations. And you have to understand, moreover, that the picture you get is, at best, only a sketch — and, in many ways, a somewhat random one. If the receivers or sensors had been placed differently, that picture, too, might be different. The seismic profile that we draw may be up to 20 metres off, either way, while the thickness of the strata we need to get to can sometimes be just five metres thick: you have to get through more than a kilometre of it, and its discontinuous, patchy, and of differing heights. Under that level of uncertainty we take a gamble on starting to develop the field, and take on considerable responsibility — because drilling, and building the necessary infrastructure, costs an enormous amount of money, which we have to recoup. So we always succeed — or else we simply don’t proceed with that project.
But we’re always trying to improve our models, making them ever more accurate in order to constantly reduce our costs. A large part of the data we obtain is circumstantial, and its important to interpret this correctly, looking for correlations, and for that it’s important to use machine learning, which can work with massive volumes of data, the analysis of which is way beyond human beings. We already have in place our own independently developed tools, based on AI (artificial intelligence) technologies. These allow us to select the best field development solutions (in terms of both reservoir geology and economics), and predict the right drilling trajectory to make sure we stay within the productive strata.
— You’ve got some extensive scientific experience under your belt. Are you currently involved in any research activities yourself?
— Yes, but solving these tasks happens within a team with distributed functions. I used to do everything myself, but I’m no longer directly involved in calculations now. My co-authors and I analyse problems, formulate objectives, and only then proceed to analysing the results of calculations performed by our young colleagues.
— Is it true you’ve even automated the monitoring of academic publications?
— It is, and this is a further opportunity for me not to leave academia. I set up a profile with my interests, and a bot sends three to five articles from various journals, or conference presentations, to my in-box every day. It means I can keep abreast of all the news, and very often what I receive serves as the staring point for new ideas, or for new developments.
I think the traditional way of working, whereby abstracts are published and patent searches are made by particular people, is wrong. Patent searches should be done by the author, particularly when submitting the terms of a patent claim — he or she, more than anyone else, is in a position to know the situation in his or her area of interest. Are you really going to entrust this to some specialist from the patent department?
— How do you rate Gazprom Neft’s technological maturity compared to other Russian and international oil companies?
— Technological maturity doesn’t mean you’re independently developing all technologies, on your own. To me, technological maturity is when you’re aware of all current technologies, understand when and how to use them, and have access to them. If you haven’t got these technologies, you find colleagues — business or research partners — to help you develop what you need. Based on my definition, we’re a very advanced company, technologically.
— And sanctions aren’t impacting you?
— All they can do is reduce the number of potential partners. They’re certainly not impacting our technological maturity because we know precisely what’s available, worldwide. And, if we haven’t got some technology or another, it’s easy to find it — we outline the task, and find companies ready to work with us in the current environment, or develop those technologies within Russia.
— You intrigued me, at the very beginning, when you alluded to new workplace management strategies. What can you change in what would seem to be as mature an industry as oil, where all roles have long been laid down?
— First and foremost, you have to introduce system engineering — analysing a field in its entirety, including its physical and economic metrics. This needs to reflect both how decisions are made, how engineering calculations are taken into account in that process, and how, at the start of a project — in the pre-investment phase, with practically no knowledge — you’re going to be able to work out the economics of that project, albeit with a degree of approximation. System engineering is important insofar as it allows you to get the maximum benefit from a project.