april / may 2016 issue 6083a7383a5e33475eed0e-e819cda5edf0a946af164bb0b2f… · ·...

April / May 2016 Issue 60

Why subsurface professionals are detectivesImproving understanding, attitudeand leadershipKeeping workforce engaged withsafetyViable plans for small reservoirsPersuading drillers to use analytics

Official publication of Finding Petroleum

Find out more and reserve your place at

www.d-e-j.comwww.d-e-j.com

Events 2014

Find out more and reserve your place at

www.d-e-j.comwww.d-e-j.com

Events 2014Transforming Subsurface ScienceLondon, 18 Apr 2016

Subsurface IT and competitive advantage Aberdeen, 04 May 2016

Transforming subsurface insightsStavanger, 10 May 2016

Finding Petroleum Opportunities in IranLondon, 19 May 2016

Finding investment opportunities in oil & gas industry companiesLondon, 07 June 2016

Finding Oil & Gas in MexicoLondon, 20 Sep 2016

Doing more with Subsurface DataKuala Lumpur, 04 Oct 2016

Doing more with Offshore Engineering DataKuala Lumpur, 05 Oct 2016

Events 2016

Find out more and reserve your place atwww.d-e-j.com

Admission

from £20

www.findingpetroleum.com

Subsurface professionals use a range of different tools and methods to try towork out where oil might be. How should this toolset be updated to take ad-vantage of new digital technology?

Ultimately, subsurface science can be summarised as “drilling profitablewells”.

The key is to learn how to do this in a predictable, repeatable way (as opposedto drilling ‘on trend’, in a pattern, or effectively randomly).

We are really detectives, subsurface detectives, taking scraps of evidence, all sorts of expert insights,and coming up with a story. The available methodologies are rule-driven interpretation; data mining(using analytics); modelling and inversion.

Rule-driven interpretationWell-established ‘rules’ have been proven for stratigraphy, structural geology, sedimentology and de-scribing petroleum systems especially by creating Gross Depositional Environment (GDE), commonrisk segment (CRS) and composite common risk segment (CCRS) maps. Nowadays these ‘rules’ aremost commonly applied through seismic data, especially 3D seismic data.

The key ‘technologies’ are a) large quantities of inexpensive multi-client 3D seismic and b) com-moditised interpretation workstations.

In truth, this methodology has now become completely commoditised. Little commercial advantageaccrues from getting it right, simply disadvantage flows from incompetent execution.

Thus, if future competitive advantage is to be found, it must lie in either data mining – applying ana-lytics to data sets that are so large that they do not allow easy interpretation by humans - or model-ling and inversion – especially those using and/or integrating more powerful geophysicaltechnologies than towed streamer 3D seismic!

Data miningWe can access satellite and airborne data, a significant variety of well results (logs, cuttings, core,flow rates), potential field, seismic, surface geology from a wide range of proprietary and publicsources in diverse formats, with different accuracy, coordinate systems, and units of measurement.

We can be confronted with truly huge amounts of data and it is critical that we extract the key infor-mation from all of it rather than looking at only a sub-set and/or simply entering the analysis with a‘going-in model’ which we then look to authenticate. Analytic techniques allow us to do this.

The key ‘technologies’ are a) the ability to integrate large quantities of diverse data, and b) fast “an-alytics” applications, tuned to the problem in hand.

Modelling and inversionPredicting physical properties such as density, magnetic susceptibility, electrical conductivity, seis-mic velocity from geophysical data whether gravity, magnetic, electro-magnetic or seismic. Also ad-dressing complex subsurface structures. Of all these technologies, seismic remains the mostpowerful, offering the least ambiguity and the most resolution, providing a framework into whichother geophysical data can be integrated.

The key ‘technologies’ are a) integration of physically diverse multi-measurements, and, b) currently‘niche’ inversion + modelling applications.

The teamAnd so, we can make a conclusion about who does this work, in a team of subsurface detectives.

Rule-driven interpretation requires seismic interpreters and geologistsData mining requires data scientists.Modelling and inversion requires geophysicists.

Integration being the key to beating the competition!

1April / May 2016 - digital energy journal

Subsurface detectivesby David Bamford, Petromall Ltd

Issue 60 April / May 2016

Subscriptions: £250 for personal subscription, £795 for corporate subscription.

E-mail: [email protected]

Printed by Printo, spol. s r.o., 708 00 Ostrava-Poruba, CzechRepublic. www.printo.cz

Front cover art by Alex McKenzie, artist, Digital Energy Journal

Future Energy Publishing Ltd39-41 North Road, London, N7 9DP, UKwww.d-e-j.com Tel +44 (0)208 150 5292Fax +44 (0)207 251 9179

Editor Karl [email protected] +44 208 150 5292

Advertising, event sponsorship and Exhibitions ManagerRichard [email protected] +44 (0) 208 150 5296

ProductionWai [email protected]

Digital Energy Journal shares expertise about how toget more value from digital technology in the upstreamoil and gas industry

David Bamford is a consultant with Petromall Ltd

www.ofs-portal.com© 2016 OFS Portal, LLC

CONNECTING THE OIL & GAS WORLD

GLOBAL SUPPLIERS.240+ OPERATORS. 25+ NETWORKS.

ONE STANDARD. THE INTEGRATED SUPPLY CHAIN HAS NEVER BEEN EASIER.


Aberdeen Subsea Expo report

Understanding, attitude, leadershipThe oil and gas industry needs to understand the current situation, improve the attitude of its leaders, reduce bureaucracyand give young people more exposure to company activities, said Chris Bird, managing director of MOL Energy UK

“As leaders, how we choose our attitude re-ally affects the industry,” said Chris Bird,managing director of MOL Energy UK. “Wecan be grumpy, happy, excited or curious.”

He was giving a talk called “transforming be-havioural change,” presented at the plenarysession of the Aberdeen “Subsea Expo 2016”event on February 3, which is organised bysubsea trade organisation Subsea UK. MOLEnergy UK is part of Hungarian oil and gascompany MOL.

Mr Bird is a former operations director ofVenture in Aberdeen, and technical directorfor Centrica UK.

Understanding the situation

The temperature in the oil and gas industryhas changed, he said. “We have to understandit.”

“The 20th century was great for oil, it will beknown as century of oil. The 21st centurywill be known as the century of renewablesand sustainability.”

“Today, there’s a myth oil is finite. No it isn't.It’s infinite. We can't use all the oil in theground today.”

“Oil money is going to be hard to come by,that's my belief. If you have that belief yourattitude changes.”

We are moving to a “VUCA” world, he said,with VUCA standing for volatility, uncer-tainty, complexity, ambiguity. This meansthat “we need to change our mindset how wedeal with things.”

“I think currently oil is at the right price. Be-fore, people paid more because they could.”

If you look at oil price data since the industrybegan in the 1860s, adjusted for inflation, theprice has only been above $40 three times,right at the start (1860 to about 1880), in thelate 1970s / early 1980s, and since the mid-2000s. “$40 is only low compared to a fewmonths ago,” he said.

Meanwhile climate change is a big threat tothe industry, with calls for a 30 per cent re-duction in consumption over the next 15-20

years, and the world to be hydrocarbon freeby 2100, which is just 84 years away.

In the past few years, the costs of oil produc-tion in the UK doubled, while the Brent fu-tures price halved.

Meanwhile there are 2.5 times more reservesglobally than there were in 1980s, and over30 billion barrels of crude oil in storagetanks. Over $400bn of projects have been de-layed or cancelled in the past few months –and a lot of that cost would have been spenton people.

“The UK Continental Shelf basin is the high-est cost basin in the world, so we have tocompete harder for money,” he said. Itdoesn’t sound like a good place to invest.

There are many changes which might happenoutside the industry, for example travel bymagnetic levitating trains could becomecheaper, faster and quicker than by plane.Solar panels could overtake hydrocarbons inour children’s lifetime, he said.

Leadership and behaviour

Companies need a mixture of leadership,management and coaching, three thingswhich are rarely found in the same company.

There are many entrepreneurial companies

with leadership and which are a bit chaotic.There are many bigger companies with goodmanagement but are very bureaucratic, withprocesses instead of leadership. “We needmore empathy, we need an open mind,” hesaid.

The industry will try to survive by putting ef-fort into technology. “We're good at that. Butwhat else?” Perhaps the industry should lookto find ways to include trust.

Perhaps it would be good if more people hadan understanding of how the oil and gas in-dustry fits into the wider picture, and what itneeds to do to make itself competitive.

Collaboration is often misunderstood in theindustry. It is a process, a way of working,not a specific thing. It comes down to align-ing your interests with other companies,choosing the right partners, and understand-ing what you know about and what theyknow about.

Oil and gas companies often do not have aculture to support collaboration, with man-agers often believing that a culture of compe-tition works better.

The industry should be finding people whocan be tomorrow’s leaders, who can work ina VUCA world, he said.

The industry is not generally very interested

We need to change our mindset on how we deal with things - we are moving to a ‘VUCA’ world - Chris Bird, managing director, MOL UK.

in these topics. “Every time we talk about be-haviour and culture it’s pushed to the side, wewant to talk about compressor availabilityand supply chains.”

“There’s a lot of ‘emotional stuff’ linked toworking with transformational change. “Mendon't like it, so we need more females here,”he said.

“It is really tough stuff, but if you do it right,we can be world class.”

Companies need to plan organisationalchange with the same level of detail as plan-ning an offshore project.

Leaders will need to develop skills in simpli-fying how they explain what the companydoes. “We mystify a lot of the time,” he said.

“We have to make decisions on intuition. Wewant strength of character.”

Companies need to make sure their culture is‘right’, he said. “How do you know your cul-ture is right? There are processes you can doto measure it. Look at blockages in the sys-tem and attack with vengeance.Deal with people, deal with culture.”

Another behavioural issue is that companiesare perhaps too obsessed with reducing risk,to the point where they actually increase riskby increasing complexity. “Our industry iszero risk tolerance. We all want to reduce riskto zero. But it actually increases our risk,” hesaid.

Reinventing ourselves

The UK oil and gas industry could go theway of the coal industry (ie nearly extinct), orit go the way of the car industry, reinventingitself and producing high value cars like theAston Martin.

“My dream is that the UK industry becomeshigh quality. We can continue to explore thefields in the UK, and be an exporter of tech-nology.”

“We have to understand the economic engine,what drives our business.We need an organisational change process,good leadership, supportive coaching,” hesaid.

“The worst thing is when managers say it anddon't believe it. We call it a 'plastic fuzzy'.”

“Ask yourself, how much time you spendeach day on 'mindfulness', thinking about

your business today and tomorrow. Do youspend your time running around like a head-less chicken? What you do today will be yourdestiny tomorrow.”

“People change for two reasons, they want tochange and we help them to learn, or becausethey've been hurt and need to change,” hesaid. “So how do we help our people to learnso they want to change?”

Why did costs go so high?

In the panel session Mr Bird was asked whyhe thought costs in the North Sea rose somuch.

“In 1998 I went offshore, we were batchcompleting 7 wells,” he said. “Today it takes3-6 weeks to drill a single well.”

“In 2008 we did 5 subsea projects in 1 year,all under £80m. Today its £180m for [one]subsea tieback. We've let our business in-crease in costs by 200-300 per cent in 5years.”

During 2008 to 2014, “everyone thought that$110 was the lowest price and we'd go to$200 and the costs would escalate.”

If you had to try to finger what is to blamefor high costs, Mr Bird blames firstly “Thecomplexity of our infrastructure and subse-quent commercial deals,” he said.

Secondly, “at the time there was a tax grab.That sent people flying.”

Thirdly, “because of $110, there was a hugedemand for resources [people]. We startedlosing a lot of our resources internally.”

One tendency the industry has is that the firstreaction to failure is to put in a process. “Webecome bureaucratic. More focussed on man-

aging the process,” he said.

Training and experience

Mr Bird was asked about whether the indus-try will come to regret the way it is currentlyreducing training budgets.

“I do a lot of work with young profession-als,” Mr Bird replied. “We want to give themtraining but we also want to give them expo-sure.”

“One guy came from downstream to up-stream and told us he needed 10 years of [up-stream] experience before he [was allowedto] do anything.”

“In my early days [in a previous oil com-pany] we took someone in their late 20s andsaid, ‘you run this big project for us.’”

“We've got to give youngsters ability to comeonboard.”

“We need better mentoring, ensure they havesomeone to talk to when going gets tough.”

This is serious

“The key thing is, is the industry in a disasteror a catastrophe?” he said.

Perhaps you can draw an uncomfortable par-allel with the Piper Alpha disaster.

Some crew members thought it was a ‘disas-ter’ went to the designated fireproof saferefuge and perished.

Others realised it was a ‘catastrophe’ and puton a life jacket and jumped. “Quite afew of them survived,” he said.

4 digital energy journal - April / May 2016


Chris Bird’s presentation can be down-loaded at www.subseaexpo.com

Delegates at the Subsea Expo Plenary Session in Aberdeen

The UK’s new Oil and Gas Authority (OGA)is putting together a clear five year plan ofwhat it will try to do. It has a list of priori-ties, ranked for urgency, said Gunter New-combe, director of exploration andproduction with the Oil and Gas Authority(OGA).

The areas of focus are put in four themes -regional development and infrastructure;technology and decommissioning; revitalis-ing exploration; and asset stewardship.There is a strong strategic focus on ‘what isbest for the UK’.

Mr Newcombe was formerly in a number ofsenior roles with BP including VP decom-missioning and HSSE manager. He wasspeaking at the plenary session of SubseaExpo 2016 in Aberdeen on Feb 5, an eventorganised by UK trade body Subsea UK.

The Oil and Gas Authority is scheduled tobecome an independent regulator in summer2016. It will be paid for by oil and gas com-panies by a levy on production.

There will be an emphasis on improving reg-ulation, including how licenses are allocated,but also ‘soft’ persuading, such as influenc-ing and promoting.

Challenges include making sure the assetsare properly looked after (‘asset steward-ship’), encouraging exploration and discov-ery, stopping the decline in production, andfinding ways to stop the cost escalation.

This means a change of focus for UK oil andgas regulation. Before OGA was formed, theUK government’s role was mainly regulat-ing oil and gas operations, but not really en-couraging it, he said.

Fighting back

“We are fighting back,” he said. “In 2015 wediscovered more than the Norwegian sectorfor the first time in 13 years, but only 13wells drilled.”

Production efficiency, the percentage of timethat oil wells are actually producing (ratherthan closed for maintenance for example),has risen from 60 per cent in 2012 to 70 percent now.

The production cost at the end 2013 was $30a barrel. “There's a real push to halve thatcost,” he said.

“Is it tough? Absolutely. There's huge effortsneeded in all those areas.”

But “there's evidence people are working to-gether and collaborating to try to recover.”

Boards

OGA is setting up a number of ‘MaximisingEconomic Recovery’ (MER) boards, to re-place the previous “PILOT” group (a part-nership between the UK oil and gas industryand government), and the Oil and Gas Indus-try Council.

There are 8 MER boards, covering explo-ration, asset stewardship, regulatory devel-opment, cost and efficiency, technology,decommissioning, supply chain and export,skills.

The boards will have representation fromdifferent industry groups, including the regu-lator, industry body Oil and Gas UK andgovernment.

Each board has 3 specific deliverables andwill meet once a quarter, and report regu-larly to government.

“The government is listening. The oil andgas industry is important to the UK, theywant to help,” he said.

“There's a real effort and focus in ensuringregulators, government and industry worktogether, attack risks and make sure opportu-nities are realised.”

Persuading

OGA sees a few times when it would bebeneficial for companies to collaborate, “butsometimes you have to force it,” he said. “Tochange behaviour is incredibly difficult.”

“Everybody tries to do the best for theircompany, understandable. But if everybodydoes that, nothing happens. You can have asmaller piece of a bigger pie, not a biggerpiece.”

“At OGA was have facilitated change bygetting people in a room and show them thecommon goal and get them to see it,” he saidduring the discussion session.

“We're the facilitator [but] it would be greatif companies could do it by themselves.”

“OGA tries our bit to facilitate but we don'thave any powers, we are using persuasion. Ithas to be invoked right now.”

Young people

Mr Newcombe was asked if he thinks the in-dustry does enough for young people.

“When you started your career many yearsago you had a lot more opportunity,” MrNewcombe said. “The oil and gas businesshas stopped doing that.”

OGA is aiming to support academia, withsome of the funding in exploration being al-located to set up a university subsurface datavisualisation facility, he said.

Safety

On the topic of safety, Mr Newcombe, a for-mer HSSE manager with BP, said, “to me,safety and efficiency are one thing. There'snothing wrong with superb safety - youshould become more efficient.”



OGA is selecting its prioritiesThe UK’s new Oil and Gas Authority (OGA) is selecting its priorities, and pushing the industry to drive down cost, said GunterNewcombe, director of exploration and production with OGA

Gunther Newcombe’s presentation can bedownloaded at www.subseaexpo.com

“Is it tough? Absolutely. There's huge efforts needed” -Gunter Newcombe, director of exploration and productionwith the UK Oil and Gas Authority



The most important factor in safety is the be-haviour of the workforce, and the workforceshould be engaged with safety, and ideally,leading safety, said Howard Woodcock, CEOof Bibby Offshore.

He was speaking at the plenary session ofSubsea Expo 2016 in Aberdeen on Feb 5, anevent organised by UK trade body SubseaUK.

Bibby Offshore is based in Aberdeen andprovides subsea construction and offshoreoperational and maintenance support. MrWoodcock started his career as a seafarer, asa deck cadet with Bibby Line, and so perhapshas more understanding of the risks of beingoffshore than many senior management.

At the moment, “a lot of people are distractedand worried about job security, [thinking]safety is maybe not that important,” he said.

“We believe further improvement in safetycan only be through engagement with theworkforce, more walking less talking.”

“Workforce engagement increases when posi-tive behaviour is seen to be the norm. Thatsounds simple, but we all have a role in this.Senior managers need to establish correct be-havioural expectations.”

Incident in 2012

A serious incident in the company in 2012prompted a lot of thinking into how to get the

workforce more engaged on safety.

A diver was working 262 feet underwaterwhen his 'umbilical' line supplying air andheat became severed, leaving him with justan emergency air tank.

It took 38 minutes for colleagues to find himand pull him into a diving bell, by which timehe had fallen unconscious. He since made afull recovery.

“We could have just investigated it andshared root cause analysis. But we decided touse this story to drive workforce engage-ment,” he said.

Bibby commissioned a professional mediacompany to produce a 44 minute documen-tary on the incident, with the length of thevideo similar to the time the diver was under-water.

Bibby provides the video in a large blackbox, together with a box of tissues, “becauseyou will feel something emotional,” he said.

Bibby wanted to use the video to build amore emotional connection with its staff,using the power of stories.

Measuring engagement

Bibby embarked on a program to try to meas-ure how its workforce feel about safety. It puttogether a survey, posing questions like,whether they would feel comfortable chal-lenging senior management.

It took several months to complete the sur-vey, with all crews completing it, collecting22,000 data points altogether, he said.

Based on the results, the company scored it-self on average ‘level 3,’ meaning that ‘work-force is 'routinely engaged with safety effort'.

There is a higher score available, for whenthe workforce is ‘leading’ on the safety effort,he said.

The survey found that every single worksitecrew was different. “It gives you specificfeedback and information on the areas of en-gagement,” he said.

Using the information from the survey, thecompany could develop unique improvementplans for each worksite.

An interesting result was that “workforcethought that senior management support waspoor,” he said.

As an example, consider whether an individ-ual at a worksite really does have the confi-dence to ‘stop the job’ on seeing somethingunsafe.

As a manager, “everyone says, ‘it is OK foryou to stop the job,’” he said. “But the work-force think, ‘do they really mean that?’”

“[As managers] how well do you understandthe way your people feel? Do your people be-lieve your story?”

“We had a discussion with teams onboardand did an improvement action plan,” hesaid.

As a result, “our statistics have improved, thefeedback has improved, maybe our work-force thinks we listen to them.”

“We reached into our extensive toolbox ofinitiatives.”

As a result, “we’ve changed the way we dothings, our divers have changed the way theydo things.”

To build confidence that senior managementmean what they say, “I personally speak toeverybody who gets injured and discuss whatwent wrong,” he said. “You'd be surprisedhow honest people are.”

“It is not a witch hunt, it is to demonstratesenior management are interested in whypeople got hurt.”

Bibby Offshore “spent years and hundreds ofthousands of pounds developing the tool,” hesaid. It is now sharing it with the rest of theoil and gas industry free of charge.

Mr Woodcock chairs the Subsea UK safetyleadership forum, where all managing direc-tors share information.

How Bibby Offshore keeps the workforce engaged with safetyIn the current business environment, many people might be distracted about job security and not thinking about safety, saidHoward Woodcock, CEO of Bibby Offshore. Here’s how his company keeps them engaged

Using a video about a serious incident to help get work-force more engaged with safety - Howard Woodcock,CEO, Bibby Offshore

Improving safety “isn't going to happen onits own,” he said. “There are lots of tools tohelp you. Share your experiences and storiesinternally and externally.”

“If we engage with the people who work forus, we’ll quickly start to reap the benefits.That takes leadership.”

Training but no commitment

In a discussion session following the talk, MrWoodcock said that the industry needed tosolve the problem of employers being reluc-tant to pay for training, if the possibility ex-ists that the staff member might leave foranother company, and “repay that investment

to another employer.”

“We have to find a way of getting aroundthat,” he said.

“We have the knowledge and ability to solveall of these problems, it is within our capabil-ity. It comes down to the will and commit-ment of individual companies.”

But “the action coming out is slow to mani-fest itself.”

Cost inflation

Mr Woodcock was also asked why he thinksNorth Sea costs went up so much over 2008

to 2014.

“I blame engineers for a lot of things. Butwe've been building cost on cost for best partof the decade,” he said.

“We didn't make it stick in the 1980s. Whatchance do we have now?”

“The common them for me is leadership andengagement. As leaders we have the power Iwould encourage everybody to think abouthow they engage.”



Developing small pools - Matt NicolIf the industry could reduce the cost of developing small discoveries, or ‘small pools’, it could make a massive difference tothe overall viability of the North Sea, said Centrica’s Matt Nicol

There are 210 ‘small pools’, discoveredreservoirs of between 3 and 15m barrels ofoil equivalent, in the North Sea, said MattNicol, director, Production & Non-OperatedAssets at Centrica (previously UK opera-tions manager for CNRL).

“These are not exploration plays, they arediscoveries,” he said, speaking at the plenarysession of Subsea Expo 2016 in Aberdeen onFeb 5, an event organised by UK trade bodySubsea UK.

A study has been carried out into what couldhappen if the capital cost of developing‘small pools’ could be reduced.

It found that a 50 per cent capital cost reduc-tion would lead to a big reduction in theminimum economically viable field size, and150 of the 210 ‘small pools’ would be acces-sible. This would lead to a billion barrels ofnew resources in the North Sea.

By comparison, 100m barrels of new re-sources were discovered during 2015 which

were in larger pools.

It would also make $20bn of investment aviable proposition, and extend the life ofmany existing assets, and have positive im-pact on suppliers and employment.

The majority of the small pools are in theCentral North Sea and the Moray Firth (aninlet of the North Sea, North and East of In-verness).

The UK North Sea Technology LeadershipBoard has identified ‘small pools’ as one ofthree main technology priorities for the en-tire industry, with the other two being inno-vations in well construction, and advances ininspection methods to manage integrity ofaging facility. It has set up a small poolsgroup, chaired by Centrica and Enquest.

Hackathon

The National Subsea Research Initiative(NSRI) recently organised a ‘hackathon’,bringing people from a range of different in-dustries to come together to try to work outways to solve the problem of making smallpools viable from a subsea perspective. 160delegates from 80 companies attended, in-cluding from academia.

The solutions were divided into ‘facilities’,‘subsea’, ‘process and flow assurance’, and‘free radical’, meaning anything goes.

The solutions were put together in a report,where the solutions were split into short term

efficiency, potential joint industry projects,strategic initiatives, and long term projects.

Ideas for joint industry projects included“hot taps” (methods to plumb a subseapipeline into an existing pipeline withoutclosing it down); new connectors, and sub-sea boosting (pumping).

“Strategic initiatives” could include subseastorage and subsea processing.

Long term projects could include separation,flow assurance, biological cleaning andsome kind of ‘downsized EOR’.

The group is aiming to match technologysuppliers with where the technology can beused.

“2016 will be an incredibly important pe-riod,” Mr Nicol said. “We've identified ahuge potential opportunity, if we can bringsome of these technologies to bear.”

Bringing small pools could also bring effi-ciencies in other ways, for example the costof running infrastructure per barrel will belower if there is more production, and thecosts of decommissioning can be pushedinto the future.

Howard Woodcock’s presentation can bedownloaded at www.subseaexpo.com

A 50 percent reduction inthe cost of developingsmall pools .. wouldmake $20bn investmenta viable proposition -Matt Nicol, director, Pro-duction and Non-Oper-ated Assets, Centrica

Matt Nicol’s presentation can be down-loaded at www.subseaexpo.com

“I believe we can have an industry success-ful at $30 [a barrel oil price], but not at theway we operate today,” said Matt Corbin,head of subsea product management withAker Solutions.

When the industry wants to reduce costs, thefirst option is generally to reduce manpower.“It means you do less with less. Let's notfool ourselves this is the answer,” he said.Similarly if we reduce contractor’s rates, itdoesn’t solve the problem.

The only answer is to find ways to do whatyou currently do at lower cost, or be ‘lean’.

He was speaking at the plenary session ofSubsea Expo 2016 in Aberdeen on Feb 5, anevent organised by UK trade body SubseaUK.

Everybody thinks they run an efficient com-pany, but every time Aker Solutions assignsa new team to try to find ways to achieve ef-ficiencies, they manage to find some, hesaid. “I urge you to put a new graduate orapprentice on your 'lean' programme, theywill open your eyes.”

As well as looking at individual processes,you can also look at your entire value streamand work out how to make it more ‘lean’.

“How many project managers manage proj-ect managers?” he asked. “It works for indi-vidual companies. But as an industry, theopportunity is massive.”

Standardisation and simplification

There is also a “standardisation and simplifi-cation” subgroup, part of Oil and Gas UK’sEfficiency Taskforce.

The industry has been talking about stan-

dardisation for many years but not making agreat deal of progress. Mr Corbin found anacetate presentation on standardisation frompre-PowerPoint days. “I think we'll have toban the word at some point,” he said.

Perhaps the word ‘simplification’ is moreuseful, and for people to focus on short termquick wins, he said.

The group identified how much the costs ofcertain items such as valves and fabricationvary, and found that very similar items fromdifferent suppliers can cost 70 per cent more.

A great deal of money is spent on documen-tation, inspections and qualification

A lot of this could be cut for example if theindustry all agreed on the same specificationfor paint, or it avoided the need to trace sim-ple items such as fasteners.

There is too much bespoke engineering inthe subsea industry, which is partly as a re-sult of people’s “drive to engineer”, he said.

It all leads to much more documentationthan is necessary. One item can have a 200page document, where the first page is a‘certificate of compliance’, the other 199pages are all the details.

Centrica’s Matt Nicol also had a view onthis, seeing lack of standardisation as onereason for the cost inflation.

“We produce a 4 inch valve today, there are27 different specifications all with 1mm dif-ference,” Mr Nicol said. “27 companies askfor essentially the same valve.” It is because“an engineer in a company considered itshould be slightly different.”

In the panel session, Gunter Newcombe,head of exploration and production with theOil and Gas Authority, also expressed a viewon this. “Everybody has their own ‘innova-tion’,” he said. “It’s not innovation, its, let’smake lots of different widgets.”

Condition monitoring

The industry should also move to better con-dition monitoring and maintenance regimesto help with asset stewardship.

As an example, consider that a few yearsago we took our cars to a garage after abreakdown. Now we are moving more todoing maintenance at regular intervals, orcars which tell us when they need to betaken to the garage. At the same time, theaverage car mileage has increased fromaround 100,000 miles to 150,000 miles.

“Why is that beyond us in subsea - why can'twe move to a proactive asset management?”Mr Corbin asked.

How to be more collaborative

In the panel Mr Corbin was asked how theindustry can be persuaded to be more collab-orative, when there is such a competitiveculture.

“I think it’s a silver bullet if we can find theanswer to that question,” he said. “We havea very commercial industry where every-body is out to win.”

“We're going to have to be willing to give upcommercial advantages and I don't think thatmindset is there. Are we moving fastenough? Absolutely not.”

Chris Bird, managing director of MOL UK,also on the panel, had a different answer,saying that people just need to try a bitharder and understand what collaboration re-ally means.

“People think it means 'we're going to befriendly,” Mr Bird said.

“But it is about alignment, selecting partnersand an exit strategy.


How to be lean - Aker’s Matt CorbinThe only way companies can survive at low oil prices is to improve productivity. How should they do it? Matt Corbin, head ofsubsea product management, Aker Solutions shared his thoughts


Matt Corbin’s presentation can be down-loaded at www.subseaexpo.com

You can look at your entirevalue stream and work outhow to make it more lean -Matt Corbin, head of subsea product manage-ment, Aker Solutions

Ikon Science’s “Ji-Fi”, or “Joint Impedanceand Facies Inversion” software, which aimsto provide a much clearer understanding ofthe subsurface, is now used by "leading mid-size active independents in US," just oneyear after the commercial launch of the soft-ware.

It is also used on two of the largest oilfieldsin the North Sea, and one of the world'slargest oil discovery in the last 10-12 years.

The software is used to ‘invert’ seismic databy creating an optimum model of rock prop-erties, or impedances, which can reproducethe observed seismic image. These imped-ances are related to the speed of sound andthe density of the rocks. With increased un-derstanding of these properties, the geoscien-tist can better detect and predict the presenceof high quality reservoir rocks, oil and gas inthe subsurface.

“I've been in this business 37 years, and Ji-Fiis as close to the holy grail [of understandingthe subsurface from seismic] as I've everseen,” says Martyn Millwood Hargrave,CEO of Ikon Science.

“The thing that makes me excited is thatwe're right at the beginning of it. The moredata you see and the more results you get, themore enthusiastic and competent you get.”

"It is going to make a big change to the waypeople explore and produce. This is an inno-vation the industry will make great use of,over the next 10 years."

"America has taken to it very quickly. Theyare quick to adopt new technologies.”

“Here [in the UK] we're mainly using it onservice projects on producing fields and help-ing to de-risk drilling prospects," he says.

Tullow Oil, which financed the developmentof the system, had exclusive access to use itfor a while. Tullow has used it on its TEN(Tweneboa, Enyenra and Ntomme) and Ju-bilee fields in Ghana."Like most people, Tullow tried it out onareas they had good well control and thoughtthey knew what was going on," Mr Millwood

Hargrave says."Now they useit in newareas."

The systemhas been com-merciallyavailable forsince early2015. "We'vegot 10 or 12customerswho weren'tusing it a yearago, nowusing it veryhappily."

The company has been working towards thisfor 15 years, and been working intently on itfor 5-6 years, with some of the company’stop technical staff, including company tech-nical director Drs Michel Kemper, describedas one of the world’s leading experts in rockphysics and inversion technology.

Partly as a result of developing the technol-ogy, Mr Kemper has been appointed an 'hon-orary lecturer' this year for the Society ofExploration Geophysicists, and will be talk-ing about it at SEG events.

There was also innovative mathematical andalgorithmic methodscontributed by the Com-monwealth Scientific and Industrial ResearchOrganisation (CSIRO) in Australia.

The software aims to make the work as un-complicated as possible. Since Ikon Scienceis a software company rather than a researchcompany, it takes seriously its role to "makethings easier, more effective and more effi-cient, so things don't break," Mr MillwoodHargrave says.

The software is available with a wide rangeof different packages, including leasing thesoftware rather than purchasing it, or pur-chasing a service.

What it is

Ji-Fi is a complex piece of software – but one

way to explain it is to say, it is pre-pro-grammed with algorithmic models of imped-ance and seismic reflection characteristics ofgeneral number of rock types.

For example, if you are looking for carbon-ates, you can do the seismic inversion on thebasis that the rock actually is carbonates. Ifyou end up with a good model, then it proba-bly was carbonates. If you don’t, then youknow it wasn’t carbonates, and you knowthat the rock doesn’t hold what you are look-ing for.

"When you are looking for rocks there’s onlycertain types of rocks that exist," Mr Mill-wood Hargrave says. "Wherever you drill inthe world you're going to find the same rocktypes, there’s only five or ten."

"In petroleum geology you're looking for cer-tain types of reservoirs, carbonate reservoirs,clastics, sandstone reservoirs- or unconven-tional reservoirs - a mixture of sand or shale,that's kind of it. There are subtle gradationsbetween them all."

"If the program finds something that 'smells'like a carbonate reservoir and has rock prop-erties of a carbonate reservoir - it probably isa carbonate reservoir."

“And if you know what you're looking for,you can reduce the complexity of the inver-sion problem."

"It is basically a way of calibrating seismic


Subsurface

Ikon Science’s Ji-Fi now used by “leading US in-depents and supermajorsIkon Science’s “Ji-Fi” or “Joint Impedance and Facies Inversion” software is now used by “leading mid-size activeindependents and supermajors in the US, and have provided valuable results on two of the largest oilfields in the North Sea”just one year after commercial launch

Ji-Fi is designed to extract geological facies and petrophysical properties from the complex seismicsignal and deliver directly a 3D description of the spatial distribution of facies and properties withina reservoir

data using rock physics, and a very sophisti-cated mathematical model," he says.

The results of the seismic inversion are de-tailed information about the rock structure,and perhaps information about fluids.

You can get a rough starting idea about whatrock you are likely to encounter from re-gional information, outcrops or nearby welllogs. The well logs don't necessarily need tobe close by - they can be from wells hun-dreds of km away.

Another way to explain Ji-Fi is to say thatyou start off with separate ‘low frequency’ orsimplified models of the subsurface, cover-ing any rock you expect to find, such asshale, water bearing sand and gas bearingsand.

This is much more powerful than developinga single low frequency model covering allphases, as interpreters were previously tryingto do.

Previously, rock physics was usually beingused as a separate process to the seismic in-terpretation.

“It is a combined problem," Mr Milton Har-grave says. “You've got a problem for rockphysics, and problem of doing the inversionof seismic data. We focus on both of thoseproblems (at the same time).

Seismic inversion

Other than Ji-Fi, there are two main tech-niques for 'inverting' seismic data to get amodel of the subsurface.

The first is 'model based simultaneous inver-sion', where you build a model of how youthink the subsurface actually looks like, andthen gradually improve it.

The second is geostatistical inversion, whichuses statistical techniques and advancedmathematics to work out what the subsurfacemost likely looks like.

The problem with both techniques is that youusually need to have a lot of wells in the re-gion under study.

Model based simultaneous inversion is a"fast technology which is good - but doesn'tgive you a reliable result unless you've got alot of wells to constrain the inversion," saysMichel Kemper, director of technology withIkon Science.

If you have a well in the region, you can di-rectly measure rock impedances with a welllog tool, rather than trying to guess or esti-mate it from seismic data.

Geostatistical inversion "also needs a lot ofwells, typically, it is very difficult to set upand only very skilled people can run it," hesays. "It can take weeks or months."

"We felt some innovation was needed in thisarea."

Benefits

With Ji-Fi, less subjective interpreting is in-volved, and less assumptions need to bemade during the process. You are not work-ing as though the rock is simpler than it actu-ally is.

The 'geological rules' can be acknowledgedduring the process, so you won't end up withsomething which makes no geologic sense(such as if you have water bearing sandabove gas bearing sand in your model.

All of it leads to a much clearer understand-ing of the subsurface. It is particularly help-ful when there are thin reservoir beds, smallreservoir pools, or complex geometry. It mayalso show up reservoir formations that youdidn't know about.

The process does not require having wells inthe region under study. If you do have wells,you can use the well data at the end, to see ifthey give a value for rock impedances simi-larly to your study.

It fits an exploration pattern around the worldwhere operators are moving from focussingon exploration to focussing on working outhow they can get more oil out of their exist-ing reservoirs at the lowest possible capital

costs.

It could also be explained as 'bridging thegap between geology and geophysics".

Because Ji-Fi is not completely dependent onwells, it can be used in frontier exploration(where no wells have been drilled before).

You can also develop a number of differentfacies models and test them. For exampleyou can do an 'amplitude vs offset' analysisand see if your actual AVO data matcheswhat your facies model would give if itmatched reality.

Ji-fi also creates a framework which could beused to add in other types of data, includingwide azimuth seismic, 2D seismic, time lapseand PRM monitoring data electromagnetics.Ikon Science is working on ways that thesetypes of data can be combined.

Case studies

The system was used over in one North Seaproject, starting with trends derived in a re-gional North Sea study, with no direct wellcalibration involved.

The results of the project matched the faciesand rock properties where they were knownfrom well logs.

Another test was a giant North SeaPalaeocene oil field, where Ji-Fi came upwith a result which was similar to the con-ventional methods (simultaneous inversionand geostatistics), but with a 'fraction of thetime, and involving only light work on ahandful of wells', Ikon Science says. Thework could have used regional informationinstead of well data and ended up with just asgood results.

It was used by one oil company in Ghana,and revealed significant additional oil vol-umes which have acceptable porosity andpermeability values, and which showed fluidcharacteristics which were similar to thoseseen in the main oil-bearing reservoirs.

Ji-fi was used by Woodside Petroleum off theNorthwest shelf of Australia, to help predictfluids and lithology near to existing wells,with work needed within 3 weeks.

The project was run on seismic data for Aus-tralia which is often available for open use.The oil company wanted facies models con-sistent with the wells and the geology, andinvestigate potential in open acreage.

Subsurface


Ikon Science receiving the Queen's Award for EnterpriseInternational Trade 2015. From left to right - Steve Hunt,COO, Ikon Science; Martyn Millwood Hargrave, CEO, IkonScience; Sir Ken Knight, Representative Deputy Lieu-tenant, on behalf of Queen Elizabeth II

Subsurface


For about 20 years, one of the general ideasbehind supercomputing was that as fasterprocessors came onto the market, you canpurchase them and install them in your com-puting centre, and immediately take full ad-vantage of the regular cadence of clockupgrades that came with each new generationof microprocessors, says Bert Beals, globallead, energy industries, Cray Inc.

But that doesn’t work any longer, becauseDennard Scaling no longer applies, he says.

This means that faster and faster micro-processors are now requiring more and morepower to run them, and more and more cool-ing to take away the heat.

Dennard Scaling has been one of the key the-ories behind the gradual increasing power ofcomputers over the past few decades. It saysthat you can add more and more transistorsinto the same space but with a constantpower density (amount of power per unit vol-ume).

In other words as transistors get smaller youcould get more processing capacity from thesame space without using much more power.

The theory was first written about in a 1974paper by Robert H. Dennard.

But the law stopped working around 2005 to2007, mainly due to an increase in currentleakages, leading to chips overheating as thetransistors got smaller.

This meant that if you want to increase thecomputing capacity, there will also be an in-crease in power consumption, even if you canfit more transistors into the same physicalvolume. This leads to an increase in heat andso cooling required.

This leads companies to think much morecarefully about how to put their supercom-puters together and has led to microprocessor

designs that support many cores on a singlechip versus increasing the core clockrate.

Massively parallel

This means that supercomputers must supportan architecture which is much more ‘parallel’– often with several separate machines work-ing on the same computational problem atonce.

This requires a different kind of interconnect,memory hierarchy, and input-output strategyversus a serial optimization approach,“You have to think about the overall systemsarchitecture, combined with software archi-tecture, combined with the people skills, nec-essary to deal with processing requirementsat massive scale,” Bert Beals says.

“We have to carefully design our system ar-chitectures to keep all the cores 'fed’. It isvery different from buying 1000 machines oninternet and cabling them together yourselfwith Ethernet switches.”

About Cray

Cray is a supercomputer manufacturer basedin Seattle, Washington.

The company has a performance engineeringorganization which provides expertise toCray’s customer and partners to help opti-mize their applications to take full advantageof the parallelism inherent in Cray’s plat-forms.

Cray has designed a special interconnect be-tween the different computing nodes, whichis far more efficient than Ethernet or even In-finiband. The interconnect topology is knownas ‘dragonfly’, with a direct, dynamicallyroutable connections between any node andany other node in the system.

Oil and gas supercomputing

The oil and gas industry has been using highperformance computing since the 1970s forseismic processing. That use has continuedand grown since that time. For example, it isestimated several large oil companies andseismic processing service companies haveseismic processing power that measures insingle and double digit petaflops.

“One key example of supercomputing usagewithin the oil and gas industry is the systememployed by PGS to process some of the

most complex and largest deep-water surveysever collected.

The Cray system at PGS, named “Abel” afterthe famous Norwegian mathematician NielsHenrik Abel, is 14th on the Nov 2015 “Top500” list and is the top commercial system onthe list. (The list is online atwww.top500.org/list/2015/11/)

Seismic processing

Seismic processing algorithms exist that willrequire supercomputers that are dramaticallyfaster than what is available today, Mr Bealssays.

We have requirements from the oil and gasindustry which show a need in the next 3-5years for machines that are 10x what we'rerunning on today. In the next 10-15 years,we're going to need machine capabilities thatare 100x what we're running today,” he says.

Current seismic processing methods stillmust make accommodations for the limita-tions of current supercomputing technologies.

How will we design and deploy systems tosupport the explosion of algorithmic com-plexity and massive increase in the amountand resolution of data needed to accurately,realistically model the earth’s subsurface?

To maximise drilling predictability, improverecovery ranges, and optimize production,you need as good an understanding of thesubsurface as possible. “We want to be ableto develop a highly accurate, high resolutionelastic model of the subsurface of the earthwhich is as easy to navigate as Google Earthis for the earth’s surface,” Mr Beals says. By having more processing power, geophysi-cists can do more processing iterations oftheir data before their deadlines, and the moreiterations, the better the quality of the finalresult.

Seismic shot records are also getting longer.It is not unusual for recording times to be 3 or4 times longer than shot records in the past.This leads to a tsunami of data to beprocessed.

One Cray customer has said that current fullwaveform inversion processing would havetaken ‘thousands of years’ to run with thecomputers available ‘not so long ago’, nowthat time can be reduced to days or weeks.

Cray - building a supercomputer is getting harderIf you want a really efficient supercomputer for seismic processing, you can no longer build one to handle the expansive datasets coming out of the field by just adding more processors, explains Bert Beals from Cray Inc

Showing a pathwayfor oil and gas super-computing - BertBeales, global leadenergy industrieswith Cray Inc

Data management in the downturnThe industry response to the current oil price will lead to changing attitudes to data management. Does it become moreimportant or less important? Oil and gas data consultancy NDB put together a panel of experts in London to discuss what iscurrently going on.

Subsurface


Data management in the oil and gas industry– the practise of ensuring compliance, highquality and availability of business criticaldata – was enjoying greater recognition at thepoint when the oil price began to fall. How isthis progress affected by the oil companies’response to the low oil price? What are theessential data activities for the oil compa-nies?

Oil and gas data consultancy NDB put to-gether a panel of data management experts,including people who had worked at formany years with many oil companies andservice companies, to discuss the question.This article is based on the key points raisedin the discussion. The full list of participantsis shown below.

The oil companies response has to reducecontractors reduce staff and cancel or post-pone projects. Business projects as well asany projects to address data. Typically thedata models and plans that are created inprojects that are stopped are not effectivelyarchived. Meaning that the technical recordsare lost and will need to be recreated.

However, the low oil price could mean thatoil companies take data management moreseriously, because there might be mergersand acquisitions of companies coming up,and a company with better organised datamight be more valuable to a buyer.

One of the barriers to better data comes from

a lack of engagement from the business, at alllevels. Perhaps the discipline and manage-ment teams have more time at the moment toexamine their own practice and how it isrecorded and managed (through data).

Divestment projects

The quality of the data impacts the value ofthe asset.

Divestment projects have been a majorsource of work for data management consult-ants over the past year and a half. It can in-volve a great deal of work, sometimes feelinglike untangling a ball of wool.

There was one example of a UK companywho wanted to sell a UK offshore asset, buttheir data was not in very good condition,which meant that the sale dropped through,the buyer didn’t want to take on responsibil-ity for the data.

Under UK law, companies are responsible formanaging data about their operations forever,and the only way they can avoid this respon-sibility if by finding a buyer for the asset.

There is another example of a sale of someoffshore oilfields, which was close to agree-ment. The sale included all of the seller’s as-sets in and offshore UK.

A data management consultant, assisting withthe transaction, found that the selling com-

pany had also done some onshore drilling, in-cluding on land which was later paved tomake a road leading to the company head-quarters, which the company had completelyforgotten about.

The buying company would have been re-sponsible for this forgotten onshore well –and liable if it had caused any problems.

Increase capability

Consultants – people who previously workedin the industry in full time positions but arenow available for a daily rate – could findthemselves in much bigger demand after theindustry gets going.

“You can get consultant onboard quickly andcheaply,” one participant said. “Consultants[will be] the main sources of expertise.”

However companies might be advised not tooutsource too much. One participant recalledan example at a (name undisclosed) oil majorthat went too far with outsourcing – andfound it did not have in-house personnel withthe skills to detect when a vendor was chang-ing the scope and extending deadline unnec-essarily.

The phrase ‘big crew change’ has been usedin the industry since around 2008, wherecompanies believed they were about to lose abig chunk of their skilled staff to retirement,and wanted to implement systems to ‘capture

Discussing data management in a downturn - from left to right, Dave Cowen, Steve Hawtin, Andy Thompson, Chris Eastment. Photographer and discussion facilitator, Ed Evans.


Subsurfacetheir knowledge’.Looking back on those times, it seems thatthe industry did not really address the prob-lem.

Perhaps it will be the same after the currentdownturn – the industry might lose may peo-ple and have new people entering it when thebusiness picks up – but still without the habitof getting better at capturing knowledge.

Getting people to care

A fundamental problem with oil and gas datamanagement is that staff and managers stilldon’t care about it enough. Looking after it isnobody’s job, and the people who benefit themost from it are the people who will need touse it in future, not the people who are usingit at the moment.

Perhaps the only way to get people to careabout data management is through companyenforcement. The company needs to specifythe minimum that needs to be kept in goodorder – for example say that the well logs, atleast, must be well managed.

One issue is that oil companies often seem totreat the entire IT function as a commodity,something necessary but which adds novalue, and so the emphasis is on doing it forless cost. This approach is perhaps not thebest for oil and gas data management, be-cause good data management can increasethe company’s competitive advantage.

Perhaps in low oil price era companies can bemore persuaded of the value data manage-ment can give.

“Data management is a core competency ofan oil company,” one participant said. “That’sthe only way an oil company makes money.And yet senior managers see it as IT - some-thing they want to do as cheaply as possible.”

Data management can help subsurface ex-perts get a better understanding of their reser-voirs, which becomes more critical in a lowoil price era, where there is less room forerror. One company plotted a fault 110maway from where it actually was, due to mis-aligning their subsurface data. This meantthat they actually had a lot more oil than theythought when planning the development.

Medium sized companies

Data management is perhaps a bigger chal-lenge for medium sized companies.

In smaller companies, there often just a few

people doing technical work, which knowswhere everything is. In smaller companies,there are often individuals who will do all ofthe data management as well as many otherroles. They are unlikely to be cut due to costsavings.

Larger companies, oil majors and NOCs, willtypically have entire departments coveringdata management and spend all day worryingabout it. Someone can move from Nigeria toAberdeen, and they know where to findthings. The data ‘service’ for a large companyis a relatively low cost.

It is the medium sized companies who haveproblems as they are trying to achieve a cer-tain maturity level and define a service fortechnical data. They are likely to abandonthese efforts and revert back to a reliance onone or two individuals.

Roundtable discussion

This article is based on a roundtable discus-

sion forum hosted by New Digital Businessat their office in London on January 28.

The participants were Steve Hawtin, authorof a book “The Management of Explorationand Production Data” and formerly solutionarchitect at Schlumberger;

Dave Cowen, senior consultant with NewDigital Business with over 30 years workingin petrotechnical computing, predominantlywith oil majors.

Andy Thompson, a consultant, former senioraccount manager with Schlumberger Infor-mation Solutions;

Chris Eastment, a data management consult-ant, who spent 25 years in full time oil indus-try roles, including with BurlingtonResources and Premier Oil.

The facilitator was Ed Evans, MD of NewDigital Business.

Should the recession persist or worsenthen further options such as outsourcing,offshoring, remote data storage (cloud)and ‘applications as a service’ will becomemore common.

Decisions such as these require compe-tency in E&P Data Management to siftthrough the hype, risks and potential bene-fits for such a service and to ensure thatbenefits are realised.

This leads nicely to the most importantfactor for the ‘essentials’ identified by thepanel:

Retain or build data management compe-tency within the organisation. It is criticalto the E&P organisation that data manage-ment competent resources are working inyour interest, ideally as employees.

Ensure that the formal organisation is inplace and is supported by effective Gover-nance.

Redraft your “strategy and principles”around data management, in order that allactivity is in support of the business pur-pose. This means that ‘vanity projects’ and‘cottage industries’ can be cut.

Data relating to safety critical systems ismanaged as an essential to the business.

Activity (business cases) should be exam-ined with more scrutiny. This is alreadyhappening and it is a good thing that thedecision makers look more closely at datamanagement business cases and theclaimed costs/benefits. This is high qualitybusiness engagement.

In the UK the new Oil and Gas Authority(OGA) compliance rules will allow re-duced obligations and costs.

What to stop

These are things that destroy value and in-crease the sense of division betweenDM/IT and the business.

Poorly conceived and poorly defined proj-ects and activities with unrealistic expecta-tions

Stop implementing software as ‘the solu-tion’ without understanding consequencesfor the business – including user produc-tivity, team productivity, workflows, proj-ect data and corporate asset datamanagement, support and infrastructurecosts.

Essential data activities in a downturnBy discussion facilitator Ed Evans, managing director, New Digital Business

Production operations


Zolnai - Helping companies map their processesUK company Zolnai.ca is helping oil and gas companies put together maps of their processes, which can be used tounderstand who does what at which time, and where the constraints are, and if the right information resources and softwareare in place

UK company Zolnai.ca is developing a serv-ice to help oil and gas companies understandtheir processes in more detail and put the in-formation in a diagram, so people can have abetter understanding of what they need to dowhen, and make sure the right informationand software is in place.

“Business process mapping” or “businessprocess management” is not something new,but probably needs to be done in a differentway in oil and gas exploration to how it isdone in other business sectors, such as insur-ance and aviation.

For example, oil and gas companies mighthave a small team of people, or individualsfrom a number of teams, doing a complexprocess such as mapping a play fairway, orpreparing for a licensing round.

The oil and gas industry does not work from asingle central database (as airlines or insur-ance companies might), it has many differentdatabases and file formats, which all need tobe passed from one person to another. So theprocesses cannot be managed with one mono-lithic piece of software, as we might see inother industry sectors.

Also oil and gas exploration can’t really bestructured using “Gantt charts”, which showwho is going to do what in a schedule. Theydon’t show the various dependencies in-volved (ie what different processes need tohappen in what sequence), and how the infor-mation needs to flow.

Applications

For example, the service could be used tomap the process a company follows in prepar-ing for a license round.

It could be used to map the processes in-volved in doing a “Play Fairway Analysis” ofreservoirs, to try to identify if a region of in-terest has all of the necessary components foran oil reservoir - a source rock, a charge, aseal, a reservoir. There are a number of stepswhich need to be taken to do this, all requir-ing different software tools.

It could be used to map standard subsurfacemodelling processes, and make sure every-body is aware that the master data needs to bein place before the modelling work starts.“That sounds very obvious but you wouldn’tbelieve how often people put the cart beforethe horse,” Mr Zolnai says.

Perhaps there are lessons learned from howwell the company co-ordinated its work forone license round, which would be useful inthe next one, and this can be included in theprocess map.

Andrew Zolnai says that based on his past ex-perience doing this sort of work, “half thetime people never really thought about whatthe processes are”.

The processes are nearly always unique to thecompany. “No two process maps will look thesame between different departments,” hesays.

Benefits

Process mapping can benefit all domain ex-perts in the company, including geoscientists,engineers, IT people, business people, finan-cial people, and senior managers. The processmap can be understood by both technical andnon-technical staff.

It can help people determine what is most im-portant. The tool could also be considered a‘mind map’ – it helps experts get a better un-derstanding of where they are at the moment.

The tool enables experts to spend more timedoing what they want to do – expert analysisand decision making – and less time trying tochase down the right piece of data or workingout what they need to do next.

Having everything laid out in a process dia-gram can help the company understand whereweak points are, for example a high amount

of reliance on a certain vendor, or that therewill be a certain task with a short time win-dow coming up, due to a delay to get a neces-sary government permission.

People know what to expect from their col-leagues – what work should arrive at whatpoint in the process, and what they shouldleave for the next person to do their shift.You can track the information which needs tobe shared between different disciplines, suchas engineers and geoscientists.

This process mapping can help companieswork with new staff, particularly staff fromother cultures, who may pick up differentlyon the way the company works, where thingsare not explicitly stated.

You can see clearly who is expected to dowhat at which point in the process. There ismuch less potential conflict.

Once you have the map in place, you can seeexactly what software tools specific individu-als need to do their work and understand whatis going on, and make sure they have them.

This work can help identify problems withyour information flows – for example where acertain piece of information (which might berequired later) is not being properly stored orlabelled, or if someone does not have the soft-ware they need to do a task.

This process can sometimes show up the rea-sons why some things aren’t working as wellas they should be, for example a certain stephasn’t been followed (like analysing the un-derstanding of the oilfield), which can lead toproblems (e.g. you might be working on awrong assumption about the structure of thereservoir).

Software

The core software to make process maps forthe information supply chain is produced by acompany based in Wellington, New Zealand,called LINQ Ltd. (website www.linq.it ) .

The LINQ software is designed as a tool to“bridge the communication gap between tech-nical experts and non-technical executives,”helping people to “target the most valuable ITimprovement first.”

Helping oil companiesmap their processesand information supplychains - Andrew Zolnaiof Zolnai.ca



Subsurface software company Paradigm haslaunched version 15.5 of its subsurface solu-tions suite, with an aim to help geoscientistsimprove productivity and maintain high reso-lution models throughout the interpretationand modelling process.

Version 15.5 also has tighter integration withother Paradigm products, and better connec-tivity with software from other companies.

Seismic data is increasingly being acquiredin high resolution, and when companies wantto reduce seismic costs, they are tending tocut back on the lower resolution surveys, notthe higher resolution ones, says IndyChakrabarti, senior vice president of ProductManagement and Strategy at Paradigm.

It is important to be able to work with thishigh resolution data all the way through thesubsurface interpretation workflow, other-wise there’s not much point in having it.

Oil companies are demanding more accuracyfrom geophysicists, because their drillingbudgets are lower. This means that they needgeophysicists to rank, or ‘high-grade’ theirprojects, rather than just give them a list ofplaces which look like they are worthdrilling.

“In the past, they’d say, ‘we have 10prospects, we’ll go after them all,’” he says.“Now they say 5.”

To help geoscientists work faster, Paradigmis making efforts to reduce the number ofmouse clicks required to do a particularworkflow, with a 2 per cent reduction beingseen as worthwhile.

Paradigm is also aiming to increase ‘automa-tion’ of the workflows. This doesn’t meanautomating the actual interpretation of thedata, but making the steps easier – so (for ex-ample) the data you might need next couldbe pre-loaded. This is like the way thatGoogle Maps pre-loads a map next to the oneyou are looking at, so it is available quicklyif you need it.

The computer system can also remember theprocedures you are doing. So if you aredoing a certain analysis process on an area ofyour interpretation, you could move the cur-sor to a different area, and the computer

would run the same procedures on the newarea.

Cross domain workflows

Paradigm has also been developing ‘crossdomain workflows’, where people from dif-ferent disciplines such as petrophysicists andgeophysicists can work more closely to-gether.

They can work on the same data, but it is dis-played in their respective software applica-tions the way they want to see it.

Also the petrophysicist’s updated data is im-mediately available to the geophysicist andthe subsurface modeller.

There is also better integration between thegeologic modelling and interpretation appli-cations, and better ways to add Paradigmwell display and mapping capabilities.

“Applications plug together, and data natu-rally flows from one user to the next,” MrChakrabarti says. “Now the guys down-stream don’t have to load the data”.

Paradigm and GE Oil and Gas

To help production engineers get a better un-derstanding of the reservoir, GE Oil and Gasas announced a plan to tie together its“RDPO” (Reservoir Driven Production Opti-mization) service with Paradigm’s subsurfacemodelling software.

This means that they can do their RDPO ana-lytics with a 3D model of the subsurface im-mediately available.

In the future, RDPO can run on GE’s“Predix” cloud service designed for indus-trial big data and analytics.

Production engineers – people who make de-cisions about day to day production – typi-cally only have information from specificwells to use as a basis for their decision mak-ing.

They can see the trends in production fromindividual wells, and use this to make deci-sions about adjusting chokes, artificial lift,changing an injection strategy or infill wells.

The weakness with this is that production en-

gineers are not using the available knowl-edge of the subsurface to make their deci-sions.

For example, production from one well couldbe lower because water injection is pushingoil towards another well, and their overall in-jection is up. The best move might be to donothing, rather than increase production fromthe first well as well.

Or production could be lower because wateris entering a reservoir close to this well. Thebest move might be to drill a new well at an-other point in the field, far away from wherewater is entering.

To be able to take these factors into accountwhen making decisions, production engi-neers need an understanding of the reservoir.Not a massively high resolution one (that areservoir engineer might have), and not onewith any detailed subsurface data (that only ageophysicist might understand), but just onewhich shows the key points which are usefulto a production engineer, similar to the onesdescribed above.

An analogy is the way GE has sophisticatedanalytics systems to monitor and understandwhat is happening with aircraft engines –but it realised that the value of this analyticsis limited if you don’t know anything aboutthe atmosphere around the engine – air pres-sure, temperature and quality.

There is a big prize to helping production en-gineers make better decisions. GE estimatesthat oil companies might be able to save 10-25 per cent on the lifecycle costs of operatingwells, if they can improve problem detectionand treatment design.

Production engineer decisions

A typical production engineer today mighthave 200 to 300 wells that they are manag-ing, says Indy Chakrabarti, senior vice presi-dent, product management and strategy withParadigm.

“We are adjusting the reservoir model so thatit’s relevant to a production engineer,” MrChakrabarti says. “A production engineerdoesn't need to see all the seismic and all thecomplexity of facies change”.

Paradigm: updated software, GE tie-upSubsurface software company Paradigm has launched version 15.5 of its subsurface 'Solutions Suite' - and announced atie-up with GE to give reservoir tools to production engineers

Further information is online atwww.pdgm.com/paradigm15-5

Designing a data lake for an oil and gas do-main is quite a challenge. We rarely see suc-cess stories in this domain. To extract valueout of the data is more challenging than ex-tracting oil itself.

The Data Lake is more than just data stor-age. It is a concept that defines the Data, In-frastructure, and the services that enables anenterprise to have a holistic view of their as-sets.

This covers a number of factors.

Linear Scalability/Elasticity – it should bepossible to expand the storage, and computeindependent from each.

Multi-Tenancy – data should be available todifferent user groups without actually repli-cating it, serving different uses on the mostgranular data sets.

Cost Transparency – the sponsors of theproject should have very clear expectationson what they are going to get out of their in-vestments in both short term and long termbasis.

Multi-Protocol Support – the data can beaccessed and operated leveraging multipleprotocols. A UNIX administrator should beable to work on the same data as comparedto the Windows user. Even Applicationsshould be able to seamlessly communicatewith the same data set without the need tostore a copy of the data

Security - to seamlessly integrate and applysecurity policies on the data, services and infrastructure across the enterprise. To makethe data secure where ever it resides.

Performance - to meet the expectations ofthe data users (Real-time, batch, or transac-tional). The hardware and software shouldbe optimized for every use. Every businessunit should define their performance param-eters before even starting the design of thedata lake.

Multi-Formats - provide support for vari-ous data formats and standardize it for dif-ferent types of data sets. It’s not aboutcleansing the data but to make it consistentacross the enterprise governed by business

standards and poli-cies.

Services

A good Data Lakeshould support thefollowing services.

Data and UserManagement: Toprovide seamlessservices (both forbusiness and IT) tomanage their dataand define policiesfor variety of usergroups with differ-ent needs

Data Quality: Toprovide basic qual-ity processes to en-sure the lake is notgetting dirty (Data Swamp). These processesvary from business to business depending ontheir data sources and governance around it.

Data Lineage and Catalogue: To under-stand how data traverses across the differentlayers of the enterprise from the source tothe end user/applications. It provides thedata stewards a hawk view of their data andits transformations enabling them to applyenterprise specific policies related to data re-tention, archival, access etc.

Data Analytics: To enable the enterprise torun tools and techniques (home-grown or3rd party services) to crunch the data andapply models (real time and historical) lever-aging best brains of the enterprise which isoften very limited in this domain. Businessesshould be able to leverage their 2nd mostvaluable assets (their knowledge workers) towork with the machines to extract real valuefor the enterprise. I would call human as-sisted machine learning.

Data Transformation and Processing: Pro-vides data scientists the ability to cleanse thedata, apply business specific transforma-tional rules, and business policies governedby operational norms for executing high enddata science work.

Data Access: To make the data freely avail-able to the businesses in a consistent, secureand standardized fashion enabling them towork on the data sets that concerns their spe-cific needs. The focus should be on Valuegeneration than on devising ways to extractand manage the data.

Governance: Provide the ability to managedata lifecycle, build business processes,apply policies that promote the enterprisefrom data generator to value creator basedon core principles and standards. CDO canenable the entries as a Data Driven Valuebased organization where the operations aregoverned by KPIs.

Self Service: to enable this capability theData Lake needs to be well architected basedon the key tenets defined above. It is abouthanding over the data back to the business towork on it seamlessly with minimum IT in-tervention.

Visualization: this is one of the most in de-mand capability from perspectives of thebusiness, IT and data scientist community tooperate on their data. Visual analytics andadvanced visualization is not only useful butcreates a totally new perspective of the rich-ness of information and value that the DataLake can provide.


Subsurface

How to make a ‘data lake’There has been a lot of talk lately about the benefit of putting all of your data in a ‘data lake’ so you can run analytics on it.But how do you make this approach work? Ketan Puri of Infosys shares some ideas

sresUssenisuB

e

ff DDDffeeiihhCC

civreSfleS

sOXC

n

aaataa aa eeccciifffOOaattttaaaaaDDDD

oitazilausiV s

maertSdiM

maertspU

stsitneicSataD

e

o

ee

w

ssssUUU

oDD

ssss

am

eeessssaaaCCCeee

aees rtsnw

aaa

ataD

ylanA

ihcrA

attttSSSSTI

aaDD

eaegaeniLataD

euuggoollaattaaCC

ss

stsitneicS ,

stsy ,

stttt

f

,ceti

ffa

a

Q a

C

taDytilauQaattaa

scitylanAataD

dna

y

a r

tilibapa

esUdna nrevoG

sseccAataD

eccn

f

an

ataDnoitamrofsnnaarrTT

g&

gnniisssseeccoorrPP

uM

ffffffaaatSTI

s

rrrrn

reeeeSSSSaM

ai

c

rfnItlu - stamroF

ecnamrofreP c

c ssa

v st

seeeecccciiiivvvvnemega

er

yei

rutcurts

tiruceSittlluuMM - orP

e

locoto

niL

ooo

a

iM

rfnI

ttttSSSStluM

yyttiilliibbaa /lacSrae /yyttiicitsalE

eer ee

er

-T

rutcurts

ggggaaarrrroooycnaneT

ssoCapsnnaarrTT

tsycnera

ilppA

snoitac

Subsurface


Amy Zeringue - How to optimise drilling with dataPersuade drillers to change by appealing to their competitive nature; managing the number of software packages in use;develop low cost agile tools (perhaps in-house); and have people on staff with expertise in both oil and gas engineering anddigital technology – some recommendations on how to optimise drilling with data.By Amy Zeringue, manager of information technology, Chevron Global Drilling and Completion, and co-chair of the executive committee ofSPE Intelligent Energy 2016

Start with the business problem

It’s best – it’s essential, actually - to start witha problem, described in business terms by theusers who are experiencing it, and define thebest way to address the business problemthrough a solution that incorporates improvedor refined processes, information technolo-gies (if applicable) and the right skills to en-able the solution.

From a data perspective, the biggest errorI’ve seen is that data management efforts areoften pursued from a strictly IT point of view.

We capture and present copious amounts ofdata, often overwhelming the user with visu-alizations or dashboards that don’t get to theheart of the business problem they’re tryingto solve.

Focusing in on a specific business processand breaking it down from the top – What in-formation do I need to make this decision?What and where is the data that constitutesthe required information? - is always thebetter method.

Persuading drillers to change

In Chevron, we used drilling data and analyt-ics tool to offer insights into rig and drillingperformance that had been on the decline inone of our offshore business units.

The data and the tools we presented allowedour engineers and managers to be more effi-cient in having instant access to the insightsthey needed to gauge their performance.

We then used that analysis to identify trends,behaviours, and designs that had caused thedecline, and then put in tools and processesthat drastically changed how our teams were

managed, how our rig contractors were re-warded, and how our operations wereplanned and executed.

The analysis drove our drilling departmentand our business partners to think and workdifferently. Eventually, the tool and data be-came secondary – the change was engrainedin the teams.Taking the example above, we were able tomotivate our drilling personnel to changethrough a combination of factors.

First, we appealed to their competitive nature.We compared drilling performance trendsacross rig teams and made performance im-provements visible.

Incentives helped too – things like perform-ance awards and management recognition.Performance improvement is a key focus itemfor us – if process change through informa-tion technology is a means to that end and ifmanagement supports the change, it will hap-pen.

Too much software

One challenge to new software adoption isthat we have too much.

As a major operator, we have hundreds ofsoftware packages – large and small – result-ing in redundancy and confusion.

Perhaps what’s driving the fact that the indus-try is not “taking advantage” of availablesoftware is that it’s hard to justify additionalinvestment in new tools when we have in-vested for years in multiple software pack-ages and feel we haven’t realized their fullvalue proposition.

We have implemented centralized governancefor our IT investment with Chevron’s D&Cdepartment, looking for ways to leveragecommon tool sets across our organization insupport of our key business processes.

More agile tools

Also, while the need for large scale, inte-grated engineering and geophysical applica-tion will persist, there is an opportunity for

more agile development of tools that canmeet specific business needs and offer greatvalue.

We’ve had some recent successes here atChevron with business intelligence and ana-lytics solutions that were developed for afraction of the cost of commercially availablesoftware- they solve a particular businesschallenge and have a rapid rate of delivery.

In today’s world of on-demand apps, this ap-proach offers a more immediate value realiza-tion vs. traditional major IT solutionundertakings.

What is the role of analytics

[The role of analytics] goes back to the identi-fying and addressing a business problem thatneeds to be solved.

Investing in analytics for analytics sake with-out a value proposition or having IT depart-ments shop around an analytics platformbecomes perceived a solution looking for aproblem. This is not the environment forthat.

Experts should identify where analytics canoffer the most benefit and pursue those withenergy and focus. Business sponsorship is es-sential and subject matter expert involvementis also a key to success.

One of my favorite mentors, Jim Crompton,co-wrote a book titled, ‘The Future Belongsto the Digital Engineer’.

In it, he describes the evolution of a hybridskill set, the Digital Engineer who is ‘engi-neering-talented, information-technology-competent, and business-savvy’.

Our digital world is going to insist on align-ment between traditional information technol-ogy teams and the business customers theyserve, almost to the point where the two couldeventually become indistinguishable.

Amy Zeringue is co-chair of the executivecommittee of the SPE Intelligent Energy2016 conference, to be held for the firsttime in Aberdeen from 6-8 September 2016.www.intelligentenergyevent.com

Amy Zeringue,Chevron


Drilling and Production

Oil major BP has developed a tool called“Well Advisor”, working together with ana-lytics company Tessella, which can helpcompany experts to avoid drilling downtimeand improve drilling efficiency.

Using this tool, BP’s experts have access toclear visualisations of vast quantities of real-time drilling data, allowing them to spot po-tentially unsafe and expensive issues beforethey happen, and to identify how to continu-ally improve operational processes.

High volumes of different kinds of data arecaptured throughout the drilling process by avariety of sensors which transmit data at dif-ferent intervals and often in different formats.

Parameters such as pressure, torque, hookload, weight on bit, mud flow, rate of pene-tration and RPM are captured and fed back toBP’s rig site operators and engineers.

This data can be used by BP engineers to en-sure their process operates at peak efficiencyand to respond quickly to warning signs.

For example, stuck pipe, where the drill, cas-ing or completion string gets physically stuckin the wellbore during tripping (moving thetubular in the wellbore), can cost $10,000 to$1,000,000 per day in NPT. There are alsoenvironmental and safety risks of the inter-vention needed to resolve it.

“The BP Well Advisor is one of best exam-ples of real-time data aggregation and visual-isation in the industry,” reckons MarkMundo, BP Well Advisor Program Manager.

“We have deployed this technology across anumber of challenging wells and it has al-ready saved us an estimated $200 millioncapital expenditure in reduced non-produc-tive time.”

Tessella’s role

Tessella worked closely with BP’s experts tounderstand not only the potential real-worldissues that could occur, but also the informa-tion needed to make the wisest operationaldecisions.

Tessella staff spent time understanding thework processes for drilling operations, thedata that is most valuable to engineers and

rig site operators, and the best way to presentthe information to allow them to make timelydecisions.

Part of the challenge was to develop andadapt the underlying data aggregation and vi-sualization platform so that it could best meetthe needs of BP engineers and rig site opera-tors.

Tessella worked closely with all parties tospecify new, feasible visualization ap-proaches, and associated data analysis com-ponents, in order to ensure the evolvingsystem supported BP’s workflows.

BP drew on Tessella’s rich experience ofhandling and presenting similar data in vari-ous industries to guide their thinking.

Shaping the system

Tessella specified detailed requirements, inIT and programming terms, of what informa-tion needed to be displayed and in what for-mat, what calculations needed to beperformed, and the qualities of the data feed-ing the displays.

The requirements collectively add up to thesystem features that become key to effectivedecision-making.

An example is the demarcation of real-timedata feeds into operational activities, and thefurther punctuation of data feeds with con-nection event markers, for example to indi-cate where the drill string is placed into slipsto connect a new section of the string.

Engineers are able to readily identify correla-tions and trends from the data when the visu-alisations are demarcated by suchconnections and high-level activities.

Analytical components also benefit throughthe implementation of models that are fine-tuned to the activities being undertaken.

Other examples include progress of cementjobs, pressure test assessments and rate ofpenetration analysis.

In order to improve make-up torque qualityassurance, Tessella worked together withKongsberg, the supplier of the underlyingSiteCom platform. It also worked together

with a number of Tubular Running Servicecompanies.

It was able to create a bespoke extension tothe WITSML data transfer format that ac-commodates a detailed make-up torque as-sessment data package.

As well as identifying functional features,Tessella worked to ensure that data manage-ment, configuration management and dataquality were also addressed and incremen-tally improved with each new release of theproduct.

Tessella has also worked with the platformvendor to ensure that specifications arealigned to their product vision whilst verify-ing that the final feature set met BP’s require-ments.

During software development, Tessella con-tinued to work with BP to verify that the sys-tem did what it was supposed to.

Upon delivery of the software, Tessellaplayed a key role in confirming acceptanceof its features and also in assessing its trialdeployment in the field.

Making it meaningful

BP’s Well Advisor handles hundreds of thou-sands of data points, as well as graphical in-puts such as tubular schematics, lithologiesand wellbore geometries.

Even trained experts can’t immediately spotareas of concern amongst such vast quantitiesof data.

So it was important that the displays wereclear, concise and focused on presenting keyinformation, both raw and derived, so BP’sexperts could quickly assess the situation andtake the appropriate action.

For example, when tripping tubulars into orout of a wellbore, analytical components inBP Well Advisor identify static frictionevents, break-overs, pipe stretch, and recoilbehaviors.

By observing trends in these events, BP’s en-gineers can predict problems before theyoccur.

BP and Tessella using analytics to improve drilling Oil major BP, together with analytics company Tessella, developed a “Well Advisor” tool, to help analyse and present data,which could be used to improve efficiency

Drilling and Production


Another key benefit that BP Well Advisorpromotes is global shared situational aware-ness. This means that engineers are seeingthe same information regardless of wherethey work.

Not only is the data itself robustly replicatedbetween networked databases, Tessella hasworked with BP and the platform vendor toensure replication of data configurationacross all servers.

This involved a powerful data mapping serv-ice that abstracts the variation in servicecompany naming conventions from the data

management requirements, and also allowsautomatic switching of data feeds based onprioritization rules.

This service, aligned to BP workflows, hasensured BP Well Advisor is fit for purposefor its client and equips the product for anexciting feature roadmap.

Presenting data

“The BP Well Advisor visualisations presentimportant data in a completely clear andcomprehensive way, so it’s not overwhelm-ing,” says Mark Claxton, energy sector direc-

tor at Tessella.

“This provides advisory information onwhich BPs engineers can make informed de-cisions in real time.

“It’s a bit like a car dashboard. If you seeyour temperature is too high or your tirepressure too low, you know you should pullover and do something about it.

“The BP Well Advisor is far more complex,but the concept of clearly presenting infor-mation you may need to act on, via a sum-mary dashboard, is actually pretty similar.”

Making projects more integratedImagine planning entire development projects around a certain profit margin – rather than trying to be as effective as you canand hoping you achieve a project margin at the end of it? io oil & gas consulting explains how it could workMark Dixon, COO with London consultancyio oil & gas consulting, suggested that the in-dustry should look at development projects ina completely integrated way, and design themaround achieving a certain profit margin.

He was speaking at the “Subsea Processingand Infrastructure Session” at Subsea Expoin Aberdeen on February 3.

So the work starts with the end in mind, iden-tifies the top level requirements, and thenthey can cascade down.

You can start with an understanding of theentire project, both subsurface and surface,and identify what you think are the most im-portant issues and then cascade down. Forexample, you can see which contractor com-mitments are the most important.

If there is a concern, for example differentvendors not agreeing on a standard for a sub-sea electrical connector, this model providesa basis to work out what the impact might beon the overall project. If it delays or increasesthe overall cost of subsea equipment, willthat capsize the project.

This way you can “understand how importantor otherwise any piece of the project is,” hesays.

You can see what happens if you change anypart of the project.

You can always see where you are, in termsof what net present value (NPV) you expectthe project to be worth, and what your uncer-tainties are.

You can refine your ‘master asset model’

until you have the desired business outcome– a positive net present value of the project,and a sensitivity you can live with (ie if pro-duction is at the lower end of the expectedrange, you can still operate).

“You focus on things which really matter, noton the detail,” he said.

“You can see how you generate value andwhere the break points are. As flow goes upand down, you can see what effect it has onvalue. You can make sure you hit minimumreturns.”

Above all, you “understand how your systemworks.”

This practise is known as ‘systems thinking’,and is widely used in other complex industrysectors, such as aviation and defence, to un-derstand the interdependencies and thelevers, he said.

Working with vendors

This way, oil companies and contractors canwork together, defining where they want toget to, and working with a common defini-tion of value for a particular project.

Instead of going to the market and saying“right guys, what are you going to do for us”to try to get the best price, you start off withcost as an independent variable (ie how muchyou are able to spend) and ask the supplychain what they can do for that cost, he said.

This way you can “take advantage of themarket conditions without taking advantageof the supply chain,” he said.

Uncertainty

Big areas of uncertainty come from the rateoil flows from the subsurface, and theamount of time projects will take to get on-stream, both of which have a big impact onthe viability of the overall project.

You can model what you think the varianceis, in production rates and through life costs,using error bars. “You can see how robustyour solution is,” he said.

“Robust” is a different objective to “opti-mised”, and leads to a different way of look-ing at it, he said. “Optimised” suggests thatyou work out exactly where the parametersshould be. But you can’t do this because ofthe uncertainty involved in the revenuestream, i.e. the reservoir.

io oil and gas consulting

io oil and gas consulting is a joint venture be-tween GE and engineering, procurement andconstruction contractor McDermott, aimingto help the industry move forward in the cur-rent environment, and help struggling proj-ects get going and stick to their budgets.

io used this technique in a project for a majoroperator, developing compression facility ona large gas field.

The company wanted to spend the rightamount on compression every year duringthe lifecycle of the project, taking into con-sideration the increased requirement for com-pression as the field declines, and basing thedecision-making around the target NPV.



“Turnarounds”, where offshore assets areshutdown every 3-5 years for necessary main-tenance work, don’t have a very good trackrecord of hitting targets.

Over 68 per cent of projects take more than10 per cent longer than planned, says GeoffRoberts, director energy industry strategywith Oracle Primavera.

The costs of this are enormous, when youconsider that it means that the maintenancework takes longer than expected, and the plat-form is offline, and not producing oil, longerthan expected.

Turnarounds are usually planned in great de-tail, with planning sometimes starting 18months before, and work with contractors andsuppliers – for an “execution window” of 20-30 days.

The organisation and other stakeholders (in-cluding regulators and contractors) have ac-cepted this plan, agreed that the work needsto be done, and accepted that the shutdownfor a certain period of time makes businesssense.

There have been many studies into the rea-sons for turnarounds taking longer than ex-pected, and it usually comes down to thescope of the work insufficiently defined be-fore starting, and the resources – people andequipment – not being utilised as effectivelyas they could be, Mr Roberts says.

A big cause of the problem can be poor com-munications and co-ordination between de-partments, such as engineering, operations,maintenance, procurement, and external or-ganisations such as contractors, suppliers andregulators.

In recent years, companies have got better atdeveloping and managing processes and cap-

turing data, but the overall outcome, in termsof turnarounds completed on time, hasn’tbeen improving, he says.

Project management software company Pri-mavera, owned by Oracle, is developing toolswhich aim to make scope definition, co-ordi-nation, resource utilisation management anddepartment integration much easier, he says.

Scope control

Poor “scope control” has been identified in anumber of studies as a major cause of turn-arounds taking longer than planned, MrRoberts says.

Scope control means getting a clear ideaabout what you are going to do and what youare not going to do, which everybody in-volves accepts and understands.

We have to “capture scope, challenge it, andget everybody to think about what they’regoing to do, he says.

“The ability to understand what you’re goingto do still seems to be broken. “We’re notpulling together to deliver one commonsource of scope.”

There may be differing ideas about what ex-actly needs to be done, some ideas which arenot relevant, or which can’t be fitted into the‘execution window’, the time when the plantis shut down. In which case it is better to sortit out beforehand, of course.

With good scope control, “everybody willhave a vision of what needs to be done.”

The scope discussion should start at the‘macro’ level, looking at which major compo-nents are going to see work done on them –once that is agreed, the more granular discus-sions can take place.

As part of controlling scope, you will be iden-tifying which are the best contractors andsuppliers you want to work with, and makingsure all of the necessary items can be deliv-ered on time.

Companies are also starting to think harderabout cost management. Planning discussionsused to be about 80% on schedule and 20%on cost, now they are more 50:50, Mr Robertssays.

This gives another dimension to the “scope”discussion – working out what can beachieved for the budget available.

Keeping resources working

To complete a turnaround efficiently, youneed to make maximum use of your ‘resourcepool’, which means both people and equip-ment.

Mobile computers – tablets and phones – canbe a great help here, Mr Roberts says.

You can capture ‘status’ – information aboutwhat has been done – as the work is done,rather than wait until the end of a shift to up-date systems.

This enables work planners to keep track ofhow the work is developing and be aware ofany problems. If the schedules need tochange, then everybody can be co-ordinated.Planners can also use the information do to‘what if’ scenarios, testing what might happenbased on the current status.

The mobile computers can also be used tokeep site managers and executives alerted ofanything they should know about (“manage-ment by exception”) so they can quickly re-plan.

The plant manager knows exactly where theproject is at any point in time, how the projectis going, whether the right materials are avail-able. The right documents are ready beforethe work starts, and the information is cap-tured as the work goes ahead. You can makepredictions about what is going to happen andact when things don’t happen to plan, includ-ing recasting an entire schedule.

If there is a change to plan, people and equip-ment can be put to work on another activity,rather than wait.

The displays on mobile computers can beconfigured to the needs of the person workingwith them. For example a top manager needsa ‘dashboard’ of how the whole project ismoving, and individual workers need infor-mation about their current activity.

You have the information each individualneeds, and you have the overview of thewhole project, all based on the same data.

Primavera - improving efficiencies in turnaroundsSoftware company Primavera is developing solutions to improve the efficient delivery of turnarounds of offshore assets whichshould help companies hit their time and budget targets

Helping get turnaroundwork within budget andto schedule - GeoffRoberts, director indus-try strategy with OraclePrimavera



Many oil companies are choosing satellitecommunications for monitoring onshore oil-wells, because the data communication canbe more reliable than sending it by cellular orradio, says Chuck Moseley, director of globalaccounts M2M [machine to machine] andIOE [internet of everywhere] with UK satel-lite operator Inmarsat.

Oil and gas engineers and pipeline operatorsreally don’t like even short outages in datacommunications when they are monitoringsomething important, he says.

The cost per megabyte of sending data bysatellite can be more expensive than usingcellular or radio communications, but relia-bility often justifies return on investment.

Production engineers and other people moni-toring wells from the company office oftendo not need a large amount of data.

Actually providing more data can be more ofa headache than a benefit, Mr Moseley says.Effective automation delivers necessary in-formation like changes, or threshold viola-tions, instead of large amounts of dataindicating no problems. It means that insteadof starting to think about what to automate,you start by thinking about what data you ac-tually need, who will need it, and what theycan do with it.

“I’ve seen some very extreme cases, wherepeople go from massive amounts of datawhich no-one is looking for, to just the perti-nent data which is very useful,” he says.

If you have less data, the oil and gas engi-neers “don’t have to sift through piles ofrecords and spreadsheets to get to the datawhich is meaningful to them. We’ve helpedwith the quality of data itself,” he says.

What to monitor

Remote monitoring capability is often used inartificial lift (rod pumps and electrical sub-mersible pumps) and sending measurements,for example well measurements, pressures,temperatures, tank levels and chemical injec-tion rates, Mr Moseley says.

In many cases, the most critical data is sim-ply to know if a well stops producing, be-cause this means lost revenue which isviewed as revenue which has been lost for-ever.

Or engineers might want to know if some-thing goes outside a certain range. “Usuallyengineers have a clear idea what tolerances(operating range) is acceptable, and at whatlevel they want to be alerted,” Mr Moseleysays.

A rod pump can have a device recordingpressure, temperature and the weight of therod string, and can send data if there is achange.

Rod strings can be 10,000 feet long, and havean enormous weight. If operated improperly,they ca break, causing very expensive andtime-consuming repair jobs – meanwhile,production is lost.

Similar for electrical submersible pumps, it isuseful to know if there is a change in pres-sure, temperature or flow rates, which couldindicate a problem. If the pump starts‘torqueing up’ (requiring much more force tomove the same amount of fluid), then it couldlead to damage. Or there could be a changewhich leads to a higher electricity consump-tion.

There is usually a controller on site whichlogs data continually, and then sends data bysatellite according to pre-defined criteria,such as a dynamometer card (for a rod pump)is not shaped as expected.

Data is commonly sent back at regular inter-vals (for example weekly or monthly),whether something is going wrong or not,just to give people in the office assurance thateverything is operating normally.

Companies are looking harder at remote

monitoring due to cost pressures – previouslythey may have had one field engineer per 20wells, today it could be 500, which can betime-prohibitive for personal visits, MrMoseley says.

L band communications

Inmarsat’s satellites can be accessed via verysmall terminals, you don’t need a big satellitedish.

These small terminal devices communicatedata in the “L” radio band.

The small terminals do not need specialistsatellite expertise to install or use, and drawvery low power, easily powered by batteryand solar cells.

The radio signals in the “L” are not suscepti-ble to interruption from rain, dust storms orsnow which is found in the traditional “Ku”and “Ka” band satellite communications.

Satellite vs cellular and radio

Oil and gas companies have used standardradio for data communication for many years,in licensed bands and unlicensed (free to use)bands.

But implementing radio requires a lot of in-frastructure, such as radio towers, and associ-ated cabling and planning, as well as right ofway.

The other option is cellular communications(which are used by mobile telephones). Cellular is often cheaper than satellite, and itis easy to implement, but the reliability de-pends on the availability of the cellphonetowers, and whether coverage is available.

Cellular communications often need to bemanaged regionally to keep costs down – ie ifthe data is being sent to another country, youneed to pay additional roaming charges.

This isn’t the case with satellite data, whichcan be sent anywhere you want for the sameprice. This is a helpful capability for servicecompanies operating internationally, whomight want to monitor the data from all oftheir electrical submersible pumps from oneworldwide location.

Monitoring onshore wells by satelliteSatellite communications are proving very useful in monitoring onshore wells, because of their high reliability. The cost permegabyte can be higher than cellular, but that might encourage people to optimise the information that people need to fulfiltheir tasks, says Inmarsat’s Chuck Moseley

Providing a largervolume of data fromwells can be moreof a headache thana benefit - ChuckMoseley, director ofglobal accounts,M2M and IOE, Inmarsat

Get the most from your investments in people and data. Timely, trusted, granular and integrated data on an open platform transforms the way oil and gas operators manage their resources.

What would you do if you knew?™

Discover more oil & gas industry solutions

Explore – More of Your Data. Drill – Deeper Into Your Data. Produce – Value From Your Data

teradata.com/oilandgas

april / may 2016 issue 6083a7383a5e33475eed0e-e819cda5edf0a946af164bb0b2f… · ·...

Documents