Into the Deep(water) With Data Visualization


Making decisions based on data delivered from sensors anywhere, let alone the ocean's fathoms, can be, well, as hard as fathoming the ocean. But being able to visualize that data and what it's telling you can make the situation far less daunting.

That's the hope Tom Moroney, deepwater technology deployment manager, holds out for advanced, interactive data visualizations at Shell Upstream Americas. "The visual presentation of information is very much a part of our journey," he told me in a recent phone interview.

Regular All Analytics readers will be familiar with Shell Upstream America's analytical journey from my earlier posts, Shell Taps Big-Data From Way Down Deep and Shell Cautions Patience on Analytics Adoption. Moroney, who heads a technology team responsible for providing geoscience expertise for Shell's deepwater portfolio in the Gulf of Mexico and Brazil, is overseeing the evolution of analytics-based decision making from one-off and person-dependent to automated and pervasive. His team has already seen the acceptance of event-based correlation among decision makers, and is currently working on the adoption of predictive analytics and advanced, forward-looking asset manager.

As discussed previously, SAS Predictive Asset Maintenance software (from this site's sponsor) is a critical component in its efforts to work through the data gathered from its deepwater deployments to get ahead of troubles that might affect performance. SAS Visual Analytics, Moroney said, seems a logical next step, and so his team is looking into the advanced visualization software.

"We feel data visualization is part of creating the right situational awareness for engineers and operational personnel, providing visual context for the business problems needing to be solved," he said.

An example (not specific to Shell Upstream America) of the type of visualization deliverable via SAS Visual Analytics.
An example (not specific to Shell Upstream America) of the type of visualization deliverable via SAS Visual Analytics.

Visualization already is an important feature of the company's event-based surveillance. For example, an alarm console provides a real-time look at all the active alerts across Shell Upstream America's deepwater portfolio in the Gulf of Mexico and Brazil. From a collaboration center, engineers and operations managers can instantly see where the big events are and what their impact might be, said Moroney, explaining that his team subscribes to lean production concepts. In addition, the team tracks how much value was protected or added by analyzing the number of alerts and time to resolution.

With the intent of extending the scope of its data visualizations, Moroney said the team is working with SAS and looking at its approach to visual analytics and event-stream processing. "We're beginning to use some fairly intense statistics and we don't want just to be looking at events but patterns and the emergence of patterns in real-time."

Being able to spot patterns as they emerge will be "extremely powerful for the business," said Moroney, explaining:

    Moving beyond picking up events and then leaving it up to humans to string those events together and make sense of them will help us get to that a ha! moment more quickly. If we can, through the use of statistics, begin to piece events together and get trends and patterns and present them visually, we see huge potential in terms of the kinds of decision making and diagnostics, and the timeliness of that decision making and diagnostics, we can enable.

How would real-time data visualization change how your organization makes decisions and diagnoses problems? Share your pie-in-the-skies below.

Related links:

Beth Schultz, Editor in Chief

Beth Schultz has more than two decades of experience as an IT writer and editor.  Most recently, she brought her expertise to bear writing thought-provoking editorial and marketing materials on a variety of technology topics for leading IT publications and industry players.  Previously, she oversaw multimedia content development, writing and editing for special feature packages at Network World. In particular, she focused on advanced IT technology and its impact on business users and in so doing became a thought leader on the revolutionary changes remaking the corporate datacenter and enterprise IT architecture. Beth has a keen ability to identify business and technology trends, developing expertise through in-depth analysis and early adopter case studies. Over the years, she has earned more than a dozen national and regional editorial excellence awards for special issues from American Business Media, American Society of Business Press Editors, Folio.net, and others.

Midmarket Companies: Bring on the Big Data

The "big" in big data is no reflection of the size of the organization embracing its potential.

Push Yourself to New Analytical Discoveries

Take inspiration from Christopher Columbus as you pursue your analytical journeys.


Re: Visited SAS
  • 10/7/2013 10:06:35 AM
NO RATINGS

@SaneIT, let's hope we've come a long way in the last three years in not only the trust placed in sensor data but in our ability to make sense of it. Deepwater decision making, especially with disasters like that hanging over head, have to be daunting irregardless of what the data or the experience is pointing people to. 

Re: Visited SAS
  • 10/7/2013 7:46:28 AM
NO RATINGS

I have to agree that not many companies out there are ready to act immediately on any data.  If we look at issues deep under water for an example we just have to look back at the Deepwater Horizon oil rig explosion we had in the Gulf of Mexico a couple years ago.  There were warning signs but they were partially ignored because the sensors weren't trusted.  From the investigation "The report found that on 20 April 2010, managers misread pressure data and gave their approval for rig workers to replace drilling fluid in the well with seawater, which was not heavy enough to prevent gas that had been leaking into the well from firing up the pipe to the rig, causing the explosion."

Re: Monitoring sensors
  • 10/5/2013 5:52:35 PM
NO RATINGS

SRS1, I suppose we could consider it a bit ironic, then, that the work Tom and his team is doing is aimed at getting ahead of things that would affect the performance of Shell's deepwater assets via predictive maintenance. Mind you, Tom & I never talked about the analysis of sensor data for safety purposes. I don't know if that's something that comes under his purview. 

Monitoring sensors
  • 10/5/2013 12:03:44 AM
NO RATINGS

Monitoring sensors in the ocean are a great idea but if any situation should arise that sends incorrect information due to faulty sensors, we might not know until it is simply too late. 

Re: Visited SAS
  • 10/4/2013 9:01:25 AM
NO RATINGS

I'll add this. At this point, I'd rate the interactive capability actually more than the real-time  data feed as more important. The interactivity can deliver instant insight, and while getting that instant insight on real-time data is great, I'm not sure too many organizations are yet prepared to act on it in real-time. Would you agree?

Re: Trusting the data
  • 10/4/2013 8:33:55 AM
NO RATINGS

SaneIT, so... the knowing you can't easily get to the sensor or manually check the environment looms over the data collection process regardless of whether or not it's going smoothly at the time. That actually seems a reasonable cause of doubt, which hopefully could over time be eased by positive results. I didn't get a chance to ask Tom what happens when a sensor fails, or about the lifespan of the sensors and the equipment they're on. He did say that he did wish they'd put more sensors out there than they did. But then again, that would have come at a cost and there's always that to weigh. 

Re: Trusting the data
  • 10/4/2013 7:10:34 AM
NO RATINGS

I think it's a more general problem but but that having them at the bottom of the ocean makes verifying the data a little more complex.  Having spent time working in a chemical plant I can say that sensors go bad without warning but in some cases it is very easy to tell that the sensor is giving a bad reading.  For instance say you have a sensor monitoring water pressure in a long pipe line.  One sensor is showing a much lower pressure than sensors on either side of it 1/2 a mile a way are.  It's pretty easy to drive out to where the sensor is located and walk the pipeline to see if any fittings are leaking.  Or you're monitoring temperature and humidity of a room and one day the humidity reading spikes for an unknown reason so you walk in with a hand held unit and check the temp and humidity.  I may be wrong here but I don't think the checks are that easy at the bottom of the ocean so they face a unique problem.

Re: Visited SAS
  • 10/4/2013 7:06:43 AM
NO RATINGS

@ Lyndon Adding interactive visualization to the mix would give amazing results. Think of the endless scenarios you could potentially generate. I can't wait to hear more about what is currently available in this area.

Re: Visited SAS
  • 10/3/2013 5:10:34 PM
NO RATINGS

..

Phoenix writes


Visualization really gives a new meaning to numbers. When you are just looking at numbers your ability to recognize trends and patterns maybe limited, since then the view point will most likely be based on logic alone. But visualization adds a new dimension and enables us to instantly recognize patterns that could have taken much longer to identify. If you add real-time visualization to the mix you will have a much faster response time. A negative result pattern could be instantly corrected.


 

I agree -- pattern/trend recognition is a major benefit. I also agree with your point about real-time visualization, and I'd add interactive visualization -- the ability to input alternative assumptions, scenarios, etc. and get an immediate visualization feedback of the effect.

 

Re: Visited SAS
  • 10/3/2013 9:21:44 AM
NO RATINGS

@Phoenix, exactly! I'd love to see a study or statistics on the difference in time to decision-making before and after an organization begins using real-time data visualization to deliver analytics.

Page 1 / 2   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +