As I’ve previously mentioned on Quora, Fredrickson listed what they found were the major sucess factors in getting a crowdsourcing project to work. When in crowdsourcing you put out an open call for novel ideas and solutions to a problem, you need to:

• Have a clearly defined problem
• Provide the crowd with enough information to work on
• Have the resources to answer questions from the crowd

Fredrickson illustrated this by describing how Innocentive put up a crowdsourcing project in the wake of the Macondo disaster, failing completely on all the above.  This was an extreme situation, where the few who had information was too busy managing the disaster to get involved in a crowdsourcing project, but sharing information openly is not something that comes natural in the oil and gas industry. Would this bar crowdsourcing from working with oil and gas?

PS: The videos from this year’s conference is coming online now, more on that later

In the spirit of Integrated Operations, there’s a live feed of the conference on the internet. You get the presentation up close and can flick back and forth in the slides if the presenter is going too fast. A virtual front-row seat!

A classical example at IO10 was the “pumps & pipes talk” below, which discussed what oil and gas could learn from medicine.

In principle, blood vessels and oil pipes are both tubes that transport fluid from point A to B in an efficient manner. Both of them can clog and in both cases someone has to go in with a tool to fix it. The surprising thing is that this line of reasoning works! There really are both lessons to be learned and technology to be transferred between the two industries. For one, both do difficult computer simulations of a non-newtonian fluid flowing through a narrow pipe. But then again, math underlies everything doesn’t it?

Extra points for those of you who can tell what commercial the video is inspired by.

For instance, you may want to calculate the pressure down in a well filled with a specific drilling mud. If the mud isn’t moving around and is not very compressible, the pressure is approximately p =  ρgh where ρ is the  fluid density, g the gravitational constant and h the height. We want the pressure in bar, we’ve got g in m/s² but mud density might be given in ppg or “pounds per gallon”. The experts may have the conversion factors memorized, but for the rest of us it’s a lot quicker to visit wolframalpha.com and simply write:

12 ppg * gravitational acceleration * 3000 m in bar

which outputs 423 bars as well as the value in atmospheres, pascal and psi in case you’r interested.

Similarly, when you’r extending the well you may want to know how much mud you need to fill the extra volume. We may state the length we drill in meters, but the diameter of the well depends on the size of the drillbit, which is always given in inches. So will this volume be given in cubic metres, litre or perhaps gallons? Here’s where one of the famous locals enter the scene, the “barrels of oil” (bbl) as a standard measure of volume. It’s the natural unit if you ship or store your drilling mud in oil barrels I guess. Again, the calculations are not really complicated, but the more units we juggle the larger the chance of a mistake. After all, you’r probably just doing a quick calculation as a reality-check for your computer simulation, right? Typing:

volume of cylinder with length 23m and diameter 8 inches

tells you that the extra length of hole needs 745.9 L of mud, which via 745.9 L in barrels of oil concludes with a need for 4.7 barrels of your finest drilling mud.

Wolfram Alpha seems weak on nested evaluation and won’t accept these two queries formulated as one sentence. (Or I might be missing something?) but it sure is a lot faster than looking up the conversion factors every time. So with Wolfram Alpha, you save ten minutes checking and double-checking your results or  a whole day searching for bugs in your simulation, depending on which you check first

The International Conference on Integrated Operations in the Petroleum Industry is taking place 29th-30th of september in Trondheim, Norway.  It’s a conference with invited talks covering different aspects of Integrated Operations. With talks having titles such as “Common issues in space and offshore“, “Towards autonomous control of a subsea drilling rig – an agent oriented approach” and “Industrial Gaming large scale Demonstration“, it looks like it’ll be an enjoyable conference.

The European Symposium on Time Series Prediction (ESTSP) takes place right outside Helsinki in Finland 3rd-5th of February, 2010.  This is a comfortably small conference with mostly one track, where you actually stand a chance getting to know the other attendants.  It’s arranged by the Adaptive Informatics Research Centre, a Center of Excellence at Aalto University.  The ALGODAN centre of excellence is hosted at the same department and together these centres is one of the biggest hubs for machine learning and soft computing in Europe. In and around this centre there’s groups doing such different things as bioinformatics, natural language processing, computational cognitive systems and of course there’s also a group on time series prediction.

Keeping with tradition, there’s also going to be a competition in time series prediction. For the previous conferences, this has been a great way to discuss and compare different methods.  Though I’m not sure if there’ll be papers on the prediction contest itself this year.

Deadlines:  Title and abstract by September 15th.  paper by the 1st of October.

The new logo of Statoil(Hydro)

After Statoil merged with the oil and gas division of Norsk Hydro two years ago, it changed its name to StatoilHydro and took on a new logo. This was to be only a temporary re-naming, it is now reverting to the old Statoil name and today unveiled a radical new logo. Or at least “radical” when considering it’s an oil and gas company. The magenta “star” above is, according to the press release, meant to symbolize Statoil as a guiding star or pathfinder, in the search for oil, gas and other sources of energy.

The old Statoil logo, notice the drop of oil

The logo garnered a lot of press. Some see a windmill or a water turbine in the logo. While Statoil do invest in renewable sources of energy, it is unlikely that the designers had this in mind. BP (British Petroleum), which also brand themselves as an “energy company” rather than oil and gas, has a logo that looks like a flower. But BP and other companies do not play with their logos. We would never see Statoil e.g. spin their logo on a pole to get the windmill association across. That would be sacrilege for any company. (Radical media start-ups excepted but again, this is the petroleum industry folks! )

I’m however convinced that a large international company like Statoil has had the design company (Scandinavian Design Group) test the logo on people from different cultures. Here I suspect the designers have done a remarkable job at meshing different messages.

The Selbu pattern is still popular

First, what the press release conveniently omits, is that no norwegian is able to see the logo without associating it with the traditional and widespread knitting pattern called the Selbu pattern. This is used extensively on traditional woolen gloves and other garment and associated with winter, skiing and a traditional way of life. A knowing wink to us from a company with the norwegian state as a majority stock holder.

The lotus flower

The logo of oil/energy company BP

But how do foreigners view “our” company’s new logo? What do you guys out there see in the magenta star? I’m only guessing but I think that people in Asia will make an association with the lotus flower as seen from above. Both the colour and the “petals” seem similar, don’t they? Statoil has been set on establishing themselves as a provider of Liquefied natural gas (LNG) in Asia and would benefit from communicating well with the emerging asian markets. As for exploration, Statoil is today present in Indonesia where it cooperates with Pertamina on the deepwater blocks Kuma and Karama. It is also involved in the Lufeng field off the coast of China.

If this interpretation of the logo is correct, Scandinavian Design Group has done a (to me) remarkable job mixing two hidden messages into quite a simple star-logo. But while the message sent by the Selbu pattern is clear, what connotations do the lotus give?

Having no asian friends to call, I’m relying on wikipedia which states that both hinduism and buddhism uses the lotus flower as a symbol of the power to create. But foremost, the beautiful lotus flower which rises from the muddy water is a symbol of purity. Now that would fit the brand of a company selling clean-burning natural gas or a company transitioning into renewable resources for that matter. But this might be taking the interpretation too far, even for this “radical” choice of logo

They've got a point about the sky

Update November 09: The guys in Statoil that worked on the logo say it’s a guiding star. Stars symbolize both energy and navigation. The half-detached “petals” apparently indicate that the logo is both a star and a constellation.  The marketing director says that the choice of colour reflects a winter sky in northern latitudes, much like the one to the right. The working vision during the design was ”Crossing energy frontiers” and one have to agree that an explorer trekking across the arctic under a magenta winter sky is a captivating image. In other news, Statoil confirmed the re-branding into an energy company, with CO2 removal as one of their stronger points.

In addition, there’s a nifty continuous plot of how the uncertainty in the forecasts on rainfall and temperature develop through the forecast period:

In this plot, reality will lay inside the dark grey and dark blue areas in 50% of cases and as far out as the light dark and light blue areas in 30% of cases. We can see how the temperature predictions get more and more uncertain further out in the forecast, a great way to visualize the butterfly effect! But the figure holds a small surprise too: The uncertainty in rainfall actually decreases further out in the prediction period! We also notice in the previous figure that accuracy in wind speed predictions picks up on the last day of the forecast.  This runs counter to our intuition, which is just why we need visualizations like these.

To be able to present the uncertainty of their models, Yr.no or the Norwegian Meteorological Institute which is behind it, had to rework their whole approach to weather forecasting. They went from running one model of the weather to running an ensemble of 51 slightly different simulations in parallell. The long term forecast is established by a majority vote between the simulations and the number of dissenting models gives the forecast uncertainty. Temperature is determined by the median value. The visualization is somewhat similar to a spaghetti plot, like in the picture to the right. (Click for larger version)

Yr also seems to take into account how well the models perform at a particular place. Some geographic locations are dominated by quick changes in the weather that are hard to predict, other places are handicapped by being far from weather stations. This is reflected as a systematic uncertainty in the long term forecast.

yr.no is hugely popular in Norway, which means that many here will be competent users of these kinds of plots.  This in turn means that other professions would have an easier time explaining the uncertainty in their forecasts if they used yr’s template. Is this feasible? The yr template has some shortcomings which may not matter much for weather predictions  but is crucial in other fields. Most obvious is perhaps how the unlikely outcomes are glossed over. There could be a 10% or 0.0001% chance of a tornado but you can’t tell from the plots. This cutoff is somewhat remeniscent of Taleb’s Black swan theory and his critique related to the financial crisis. But this is also a weakness of the ensemble approach. An event that has only 1% chance of occuring has only a 40% chance of appearing in any of the 51 simulations, so low-probability events cannot even be estimated reliably.

Another weakness is that you’ll never see a bifurcation here. The plots of temperature and  precipitation assumes that the probability distribution has one maximum that tails off in both directions. The ensemble could, at least in theory have been split between say 25 simulations predicting a very hot day and 26  predicting a cold day, with no predictions in between. The plot however, would misrepresent this as a skewed distribution with “cold” as most likely and a wide tail up to “hot”.

Finally, the plots do not give you a sense of the joint probability distribution. Friday morning in the above plot could possibly be rainy or dry and it could be below or above freezing. But how likely is snow? The probability distributions are not independent so this can’t be answered by simply combining the two distributions in the plot.

We could of course modify the yr template slightly to address these issues, but would this confuse the user? Maybe anything beyond the red to green colour-coding is too much information. I’d love to hear of other ways to represent uncertainty that anyone has come across.