Oil n Gas Sensor Data + Big Data Analytics = Game Changer !!!

Each blow costs the Oil and Gas industry a Billion dollars.

Can we avoid it? Can we see it coming and take action?

We also know that each operating rig consists of thousands of sensors. This sensor data is used to analyse and reduce HSE (Health, Safety and Environment) risk considerably and dollars.

The current situation in the upstream side of oil n gas industry is that there are a plethora of fragmented data streams which have not been seen in a holistic fashion. They consist of Reserves Geospatial data, MWD Measurement While Drilling data / Remotely steerable down hole tools - RPM, Down hole pressures from fibre optic sensors, Temperature sensors, Circulation solids, SCADA data from Valve events and Pump events, Asset operating parameters, Out of condition alarms, Safety Incident data pools , Seismic Survey Data, Identity management logs ( Swipe in, Swipe outs), Contractor data points and Ambient conditions. These can be broadly segmented into 2 data classes

Velocity ( Real time streaming MWD, LWD, SCADA data streams ) and
Variety ( Unstructured Reserves data, Geospatial data, Safety incident notes , Surveillance Video streams )

So what's the cost of data fragmentation in last mile ? This fragmentation prevents risk/safety specialists from seeing the risk context holistically and the right lenses are not in place to get the context and triangulate early warning patterns.

See the full blog on Oil n Gas Big Data Use Cases at

http://blog.fluturasolutions.com/2013/10/upstream-sensor-data-big-d...

1 member likes this

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Dr. Z on October 29, 2013 at 8:41am: I think sensor data of this magnitude could be useful in other fields as well, e.g. a large boat, a plane, etc. Why focus on the Oil industry which has already passed its peak?

Comment by derick.jose on October 16, 2013 at 8:51am: Thats a very great formulation Vince !

Comment by Vincent Granville on October 16, 2013 at 8:41am: I remember that when I did my PhD, my boss used to work with statistical models to detect oil fields by drilling as few wells as possible. If you could drill 10 wells, 5 producing oil, five not producing oil, and assuming the shape of the oil field is a convex domain, then the problem consisted of estimating the border of the domain (oil field), based on 5 outside and 5 inside points, and then drill wells with a very high probability of finding oil.