Category Archives: Uncategorized

How To Turbocharge Oil & Gas Analyses With Machine Learning and The Right EIM Foundation

It is generally accepted that good analysis of oil and gas data results in actionable insights, which in turn leads to better profits and growth. With today’s advancements in technology and processing power, more data and better analysis are easily achievable but will require the right EIM (Enterprise Information Management)  foundation to make “all” data available and “analyses-ready”.

The evidence of those analytics are clear and ubiquitous. In an article in JPT (Journal of Petroleum Technology) by Stephen Rassenfoss, “Four Answers To the Question: What Can I Learn From Analytics?”, Devon Energy concludes it is possible to increase production by 25% by drilling the lateral toe-up in Cana-Woodford Shale. Range Resources, responding to a different question and with Machine Learning (ML) analysis, concluded more production in the Marcellus is associated with wells fracked with as much sand volume as the reservoir can handle.

All Data All The Time = More Studies More Return

Looking closer at the article, both studies were based on a relatively small data set; Devon Energy and Range Resources only used 300 and 156 wells respectively.  Both companies stated that a larger data set would help their respective studies. So, why some studies rely on a small population of wells when there are thousands more that could have been included to reach a deeper understanding.

While the answer depends on the study itself, we find two key data”preparation” problems that may contribute to the answer a) data findability/ availability b) data readiness for analyses. In some E&P companies, data preparation can consume over 50% of total study’s time. This is where I believe EIM can make a difference by taking a proactive role.

 Three Strategic EIM Initiatives to Turbocharge Your Organization’s Analytics

Information preparation for exploratory analytics like the above, require Oil and Gas companies to embrace a new paradigm in EIM. The traditional “data management” has its applications but can be rigid and limiting because it requires predefined schemas.

We share our favorite three EIM strategic initiatives to deliver  more, trustworthy and analyses-ready information:

  • Strategic and Selective Information Governance Program – A strong data governance model ensures data can be trusted, correlated and integrated, this is a foundational step and will take standardizing, and mastering key entities and attributes.   Tip: key enabling technology is Master Data Management (MDM)
  •  Multi-Stream Data Correlation – Together with the MDM, “Big Data” technology and processes enable the inclusion and further correlation of data from a variety of streams, without the prejudice of predefined data schema.
  • Collaborative Process and Partnership – From years of lessons learned, we’ve noticed that none of the above will move the needle much at all if implemented in isolation. A collaborative process with the sole purpose of fostering a close partnership between IM engineers/ architects, data scientists, and the business, is what differentiates success from failure. As the organization finds new “nuggets of insights,” the EIM team’s role is to put the necessary structure in place to capture the required data systematically and then infiltrate it into the organization’s DNA.

New analytics are positively changing how we produce and manage oil and gas fields. Companies that invest in getting their EIM foundation right will lead the race among its competition.

Disclosure:

For help on defining and implementing EIM strategy please contact us.
With Petroleum Engineers, Geoscientist, Data Scientists and Enterprise Information Architects on the Certis team, we help companies design and implement EIM solutions that support their business goals. for more information on our services please email us at info@certisinc.com.

The Rise of the Oil & Gas Analytical Citizen (and company)

 

Then

Fifteen years ago, while at the Society of Petroleum Engineers (SPE.org) conference, I was introduced to artificial intelligence (AI) tools specific for Oil and Gas use.  I was very excited to learn more and build models to optimize production and understand its key influencers for example.  I was certain data-driven insights were what this industry needed. What engineer wouldn’t want to use this?

To my surprise though, only a handful of engineers were ready to embrace the technology, and most said their organizations simply weren’t ready for it.

Now

Fast forward to 2017. Data-Driven and AI analytics are reasonably commonplace among engineers. Tools are found in nearly every company – not just the major companies, but also in the independent players and ambitious smaller companies. How did this happen?

This is what happened: Time, technology and people changed.

A widespread of technology is usually a result of ease-of-use, reliability, and usefulness. One needs only look as far as Apple’s iPhone. Apple created an amazingly intuitive, reliable and useful phone, with an ever-growing market of applications.
With each new iteration, more and more people wanted an iPhone. Today not only is every citizen using a smartphone but they are entirely comfortable asking digital strangers named Siri, Alexa or Cortana for directions or to plan their daily activities.

Advancements in smartphones (and subsequent widespread adoption) raised the technological comfort level of the everyday user. Consequently, today’s oil and gas citizens easily embrace new technology and will take the time to experiment with different ideas and tools.

These same consumers are not afraid of change – they expect it now.

Statistical and AI based analytical tools were (and are) perfectly placed to succeed in Oil and Gas. Increased adoption was inevitable. But they are still not at the level I expected 15 years ago. Why?

What needs to happen in Oil & Gas next?

The problem is that while the market is ripe, oil and gas infrastructure and culture must catch up as well. More integrated and better quality data must seamlessly flow to analytical tools so an average company-citizen (and not IT) can easily explore any data, trust it and generate meaningful calculations or reports, faster, efficiently and more insightful than ever.

That vision translates to three actions:

  • Prepare a data strategy, architecture, and governance that enable an analytical company.  Few advancements in the MDM and Data Lake areas that will put you on a good pathway.
  • More intuitive and easier to use analytical tools must infiltrate the organization, the way outlook or excel does. Take advantage of smart searches, NLP (Natural Language Processing), and machine learning to name a few.
  • Create and encourage a culture that expects and enforces data-driven decisions across the entire company, for this you will need a clear vision and commitment from the leaders.

Until then, AI and Data Driven analyses remain in the hands of the chosen few ‘nerds’ – thanks to The Bing Bang Theory, being a ‘nerd’ is totally cool.

For greater clarity on your position, contact Esta Henderson – esta@certisinc.com – tel: +1.281.674.3224 to schedule a complimentary strategy appraisal with Fatima Alsubhi, our CEO.

Crossing The Border From a Mere Change to Cultural Expectation for QUALITY DATA

Culture sets certain expectations of behavior, and once accepted, there is no deviation. Even if you are removed from the cultural origin, these behaviors are ingrained and follow long after. 
I recently experienced this first hand, when a dear friend of mine was diagnosed with cancer. Of course, I was very distraught. When, a few weeks later, he was admitted to hospital for surgery, I visited him and his wonderful wife. This was natural to me, visiting a sick or injured friend at home or in the hospital, is not only a kind gesture, but is an expected social obligation ingrained in me since childhood.  What seemed to my friends as a thoughtful gesture was something I could not imagine not doing, or imagine friends of my culture not doing for me.
 
It made me wonder what makes a behavior “culturally” accepted and ingrained? When did visiting a sick friend become more than a thoughtful gesture, and cross the barrier into social obligation? How did this transition occur?  
 
These musings extended to Oil and Gas corporate culture. What behaviors were so ingrained at work that they had become second nature? Did they serve a purpose, such as to improve data quality? If not, what would it take to weave in these behaviors, and make them the expected social norm, and a clear moral obligation or expected practice within an organization? In an ideal world, these cultural obligations would lead to employees and employers alike feeling that it is “on them” to report and correct data quality issues, no matter at what point in the process they were discovered.  
 
 I thought it might be a good idea to ask my readers these questions. Are such behaviors ingrained in your workplace deep enough to be considered cultural? How would you weave them in, if not? If they are a part of your corporate culture, can you point to any policies and practices that may have led to this?  

Reminded Again, Narrow Focus Leads To Failure Every Time. Why do Some Data Projects Never Make It?

toronto
In 1993, an incident occurred in the Toronto Dominion Bank Tower that caught national attention, enough so that it made the infamous “Darwin Awards”. A lawyer, in an attempt to demonstrate the safety and strength of the building’s windows to visiting law students, crashed through a pane of glass with his shoulder and fell 24 floors to his death. Maybe the Glass did not break but it pulled off the wall. 
 
The lawyer made a classic mistake, he had focused on one specific thing to the exclusion of the big picture. If he had taken a look at his hypothesis from a wider angle, he might have considered the numerous other factors  that may have contributed to his doomed demonstration – the bond between the glass and the frame,  the yielding effect of material after repeated tests, or simply the view of the courtyard below (the high risk should it fail) might have been enough to make him reconsider his “leap of logic”. He focused on a specific item and ignored the other factors. 
 
Such a narrowed focus is equally risky to an information management project, or any project really. Although we are getting better we often focus on one thing: technology implementation and ignore other aspects.
From my experience, many factors contribute to the success or failure of information management in Oil & Gas projects. People, technology, processes, legacy data, Integration, a company’s culture, operational model, infrastructure, time constraints, or external influences such as vendors and partners, just to name a few. Each has a degree of influence on the project, but rarely will they cause the demise of the project – unless they are ignored! The key to success in any project is the consideration of all aspects, and an assessment of the risks they impose, prior to spending millions.
As an example, let’s look at survey data. How would you manage that data?
Often, companies focus on two elements:
  • Finding the technology to host the data
  • Migration of the data to the new solution
Success is declared at the end of these two steps, but two years down the road, the business has not embraced the solution, or worse yet, they continue to see incomplete surveys, a problem the new technology was supposed to solve. Failure, in this case, is less abrupt than an appointment with the Toronto Dominion Courtyard, but it is failure nonetheless.
 
More often than not, projects like the one above fail to take into consideration the other aspects that will keep data quality intact.
Even more often, these projects fail to consider external factors such as data acquisition vendors. These external vendors have their own processes and formats. If your project ignores our increasingly integrated world, and cannot cooperate with the processes, technology, and data formats of key external vendors and business partners, your project will yield very limited results and will not be sustainable. 

To achieve sustainable success in data management projects or any projects for that matter, it is necessary to consider the context surrounding the project, not just the specifics. Without this context, like the unfortunate lawyer, your project too can look forward to a rather significant fall.

To Build Fit Enterprise Solutions, Be Physical …

The British and the Americans speak the same language. But, say “I have a flat” to a British, and it means something completely different than said to an American. The former would congratulate you, and the latter would feel sorry for you. Flat in the UK means an apartment. Flat in Houston means a flat tire. The same 4 words, arranged in the exact same way, in what is ostensibly the same language, and yet either speaker would confuse their audience, if the audiences were transposed.

It is the same thing in business – if you cross different corporate cultures or even inter-organizational boundaries, industry terminology might sound the same but mean very different things. Sometimes we think we are communicating, but we are not.

Why is this a problem? Because it is not possible to build an enterprise data management solution to serve all departments without addressing variations in expectations for the same word. Especially if the term in question is one that defines your organization’s values and activities.

“Sometimes we think we are communicating, but we are not”

In the corporate world of Energy E&P, the word “completion” means different things to the different departments. If you mention a “Completion” to a Landman, he will assume you are referring to the subsurface horizon for his leases (it is more complex than this, but for the sake of this argument we need not dive into details). If a “Completion” is referenced to a Production Engineer, she immediately thinks of the intersection of a wellbore and a reservoir horizon. To a Completion Engineer, the same term means the process of completing a well after the well has reached final depth.

As organizations’ data management practice become more matured, they start to make their way towards the right of the EIM MM (Enterprise Information Management Maturity Model). Centralized solutions such as Master Data Management (MDM) are important and are designed to serve ALL departments to break as many silos as possible.

Naturally, to create a centralized solution that addresses needs across the enterprise, you must first reach consensus on how to build that solution. The solution must ensure that the data is NOT LOST, NOT OVERWRITTEN and is FULLY CAPTURED and useful to EVERYONE. What is the best way to reach consensus without the risk of losing data?

Get Physical

To answer the above question, many agree that information systems need to be built based on the physical reality to gather granular data …

By basing your data on the physical world and capture granular data as practically possible, you not only make it possible to capture all related information but also possible to report it in any combination of grouping and queries. See the example in figure 1.

Focus on Enterprise Needs and Departmental needs will follow…

I have seen systems that ignore wellbore data yet store only completions per well. At other clients, I have seen systems that take short cuts by storing wells, wellbore and wellbore completion data in one line (this necessitates overwriting old completion data with new everytime there is a change), these are “fit-for-purpose” systems.  These are not enterprise level solutions, but rather serve departmental needs.

Too often systems are designed for the need of one group/department/purpose rather than for the need of the company as a whole. However, if the needs of the whole are defined and understood, both company and groups will have what they need and then some.

Let’s look at an example to clarify this position:

Figure 1 Multi lateral well

Figure 1 Multi lateral well

In Figure 1 above, how would you store the data for the well in your organization or your department? Would you define the data captured as one well, three bores, and three completions? Or maybe two completions? One?
Depending on your department or organizational definitions, any of the above definitions could be fit-for-purpose correct. Accounting systems might keep track of ONLY one completion if it made Payroll and Tax sense. While Land may only keep track of 2 completions if the bores are in two zones. An engineer would track three completions and will be specific to one completion per wellbore. The regulatory department may want you to report something entirely different.
How do we decide the number of completions so that the information is captured accurately, yet remains useful to a Landman, Accountant, Engineer, and Geoscientist? Build based on the physical reality and stay granular.
In Figure 1, physically speaking, we see one well with three paths (3 wellbores). Each bore has its own configuration that open to the reservoir (completions). In total, this well has three different ‘Completions’,  one ‘Completion’ for each of the horizontal bores.
Accounting can query how many different cost centers the well has, and depending on the production (and other complex rules) the answer could be three but it could be 1.  Depending on the lease agreement, Landman could get a result of one or 3 completions. An engineer can also easily query and graph this data to find the three pathways, and determine each completion job per wellbore.
While it could be argued that data needs to be presented differently to each department, the underlying source data must reflect the physical truth. After all, we cannot control what people call things and certainly cannot change the lingo.

Juicy Data Aligned

juice

Around the corner from my house is a local shop selling an excellent assortment of fresh vegetable and fruit juices. Having tried their product, I was hooked, and thought it would be a good addition to my diet on a daily basis. But I knew with my schedule that unless I made a financial commitment, and paid ahead of time, I would simply forget to return on a regular basis.  For this reason, I broached the subject of a subscription with the vendor. If the juice was already paid for, and all I had to do was drop in and pick it up, I’d save time, and have incentive to stop by (or waste money).

However, the owner of the shop did not have a subscription model, and had no set process for handling one. But as any great business person does when dealing with a potential long term loyal customer, the owner accommodated my proposition, and simply wrote the subscription terms on a piece of paper (my name, total number of juices owed and date of first purchase), and communicated the arrangement with her staff. This piece of paper, was tacked to the wall behind the counter. I could now walk in at any time, and ask for my juice. Yess!

Of course, this wasn’t a perfect system, but it aligned with business needs (more repeat business), and worked without fail, until, of course, it eventually failed. On my second to last visit, the clerk behind the counter could not find the paper. Whether or not I got the juice owed to me that day is irrelevant to the topic at hand…the business response, however, is not.

When I went in today, they had a bigger piece of paper, with a fluorescent tag on it and large fonts. More importantly, they had also added another data point, labeled ‘REMAINING DRINKS’. This simple addition to their data and slight change to the process made it easier and faster for the business to serve a client. Previously, the salesperson would have to count the number of drinks I had had to date, add the current order, then deduct from the total subscription. But now, at a glance a salesperson can tell if I have remaining drinks or not, and as you can imagine deducting the 2 juices I picked up today from the twelve remaining is far simpler. Not to mention the data and process adjustment, helped them avoid liability, and improved their margins (more time to serve other customers). To me, this is a perfect example of aligning data solutions to business needs.

There are several parallels in the above analogy to our business, the oil and gas industry, albeit with a great deal more complexity. The data needs of our petro professionals, land, geoscience and engineering have been proven to translate directly into financial gains, but are we doing enough listening to what the real needs of the business are? Reference our blog on Better Capital Allocation With A Rear-View Mirror – Look Back for an example on what it takes to align data to corporate needs.

There is real value to harvest inside an individual organization when data strategies are elevated to higher standards. Like the juice shop, oil and gas can reap benefits from improved data handling in terms of response time, reduction in overhead, and client (stakeholder) satisfaction, but on a far larger scale.  If the juice shop had not adapted their methodology in response to their failure of process (even if it wasn’t hugely likely to reoccur) the customer perception might be that they didn’t care to provide better service. Instead, they might just get unofficial advertising from readers asking where I get my juice. I’d suggest that the oil and gas industry could benefit from similar data-handling improvements. Most companies today align their data management strategies to departmental and functional needs.  Unless the data is also aligned to the corporate goals many companies will continue to leave money on the table.

We have been handicapped by high margins, will this happen again or will we learn?

About 15 to 20 years ago, we started to discuss and plan the implementation of databases in Oil and Gas, in hopes of  reaping the benefits of all its promises. And we did plan and deploy those databases.  It is now no longer conceivable to draw geological maps by hand or to store production volumes in books. Also, in the last ten years, we have moved beyond simple storage of digital content and have started looking into managing data quality more aggressively. Here too, we have made inroads. But have we done enough?

Have you ever wondered why companies are still cleaning their data over and over again? Or why we are still putting up building blocks such as standards for master well lists and hierarchies? It seems to me that the industry as a whole is not able to break through the foundational stages of enterprise information management.  Because they can’t break through, they are unable to achieve a sustainable, robust foundation that allows their systems to  keep pace with business growth or business assets diversification.

Perversely, I believe this is because the oil and gas industry has been handicapped by high margins. When a company is making money despite itself, throwing additional bodies and resources to solve a pressing issue seems like the fastest and most effective solution in that moment. Because the industry is structured in such a way that opportunities have to be seized in the moment, there is often little time to wait for the right solution to be implemented.

Throwing money at a problem is not always the wrong thing to do. However, if it becomes your go-to solution, you are asking for trouble.

I would argue that highly leveraged companies have put themselves at high risk of bankruptcy because they do not invest sufficiently in efficiency and agility through optimized processes and quality information flow. For example, coming up with the most effective completion for your reservoir requires access to quality and granular technical data. This data does not just happen, it takes a great deal of wiring and plumbing work to obtain your organization’s data and processes, luckily if done right, it is a one-time investment with minimal operational upkeep.

According to Bloomberg, CNN and Oil & Gas 360 reports, during this ongoing downturn, at least 60 companies have entered chapter 11 in the USA alone. Ultra, Swift, Sabine, Quicksilver, American Energy are just a few of these highly leveraged but otherwise technically excellent companies.

Without the required behind the scenes investment, engineers and geoscientist will  find a way to get the data they need to make decisions. They will, and often do, work hard to bring data from many siloed systems. For each engineer to still have to massage data is throwing money at the problem. If the correct platform is implemented in your company, this information would flow like clockwork to everyone that needs it with little to no manual work.

WHAT COULD HAVE BEEN DONE?

We all know it is never the wrong time to make a profit. Consequently, it is never the wrong time to invest in the right foundation. During a downturn, lower demand creates an abundance of the only resource unavailable during an upturn – time. This time, spent wisely, could bring huge dividends during the next upswing in prices. Conversely, during a period of high prices, it is the other resources we cannot afford to waste. During a boom, we cannot ignore building sustainable longterm data and process solutions the RIGHT way.

It is never the wrong time to make a profit. Consequently, it is never the wrong time to invest in the right foundation.

Of course, there is no single “right way” that will work for everyone. The right way for your organization is entirely subjective, the only rule being that it must align with your company’s operations models and goals. By contrast, the only truly wrong way is to do nothing, or invest nothing at all.

If your organization has survived more than ten years, then it has seen more than one downturn, along with prosperous times. If you’ve been bitten before, it’s time to be twice shy. Don’t let the false security of high margins handicap you from attaining sustainable and long-term information management solutions.

Here are some key pointers that you probably already know:

      Track and automate repeatable tasks – many of your organization’s manual and repeatable tasks have become easier to track and automate with the help of BPMS solutions. Gain transparency into your processes, automate them, and make them leaner whenever possible.  

   Avoid Duplication of Effort – Siloed systems and departmental communication issues result in significant duplicated efforts or reworks of the same data.  Implementing strong data QA process upstream can resolve this. The farther upstream, the better. For example, geoscientists are forced to rework their maps when they discover inaccuracy in the elevation or directional survey data. These are simple low hanging fruits that should be easy to remove by implementing controls at the source, and at each stop along the way.

  Take an Enterprise View –  Most E&P companies fall under the enterprise category. Even if they are a smaller player, they often employ more people than the average small to medium business  (especially during a boom) and deal with a large number of vendors, suppliers, and clients. Your organization should deploy enterprise solutions that match your company’s enterprise operations model. Most E&P companies fall in the lower right quadrant in the below MIT matrix.

mitopmodel