05 May 2011
Yogi Schulz, gives a candid interview on how the petroleum industry could do better business by actually paying attention to its own data. Part 1 of 2.
We sat down with Yogi to explore his opinions on how data quality affects petroleum companies. In his engaging conversational tone, he didn’t pull any punches as he offered plenty of food for thought.
What are the business consequences of poor data quality in the oil and gas industry?
Dramatically, the consequences are dry holes, excessive operating costs, and failing to maximize ultimate recovery from reservoirs. We can drill into that, no pun intended. How do dry holes occur?
Some dry holes occur because of survey problems. In this case you're not drilling where you think you are drilling because you have disconnects and busts in your survey data. You've got a location problem.
Other reasons for dry holes are poor seismic and geological interpretation. Sometimes these are a consequence of inadequate data management of the data required for the interpretation.
Another reason, through poor well planning, is abandoning the well before you reach the target zone, or missing the target zone.
If you don't have a good handle on your cost data, you mislead yourself about continuing to operate a well that should be suspended and probably abandoned.
How often do you see these problems?
That's the tricky part. Nobody ever publishes a press release about a fiasco like that. Many oil companies, as other companies in other industries, exhibit a lack of candor about admitting these problems occur. Well-performing companies are brutally frank about analyzing these dry hole outcomes with a view to improving organization performance to reduce the risks and minimize the probability of a reoccurrence.
I'm aware of one client where a whole a bunch of data magically disappeared after a high-cost, high-profile exploration well was a duster. It was pretty clear that the reason for the duster wasn't the lack of geological opportunity. It was due to poor data management. By literally shredding the data that would support a witch hunt, somebody saved their political hide. That's not best practice performance of an organization.
For decades IBM never had the best computers, but they always had the best marketing program. A component of that was ruthless loss analysis. When they lost a piece of business they scurried back to their office, despondent, but determined to understand what went wrong in that sales cycle. Very few oil companies do the same thing when they drill a dry hole, lose mineral rights to a parcel of land, or fail to improperly manage their seismic data.
In those cases it seems like data management is like the physical sciences. The more reliable variables you have, the more reliable predictions you can make.
We’re now at the point where the number of variables that must be juggled exceed the ability of even the brightest engineer, geologist, or geophysicist to manage. The use of software and data to produce better sub-surface models that offer better predictability and history-matching overcome our human limitations.
What about excessive operating costs?
If you don't have a good handle on your cost data, you mislead yourself about continuing to operate a well that should be suspended and probably abandoned. On the other hand, the reverse is true: you can, in the absence of good data, abandon a well that still has more remaining economical, producible reserves than you believe to be the case.
These are situations where you're being misled:
- invoices being coded to the wrong cost center;
- production volumes being allocated to the wrong well;
- losing the reservoir models that you need for superior reservoir exploitation.
Are those regular mistakes within the industry?
They happen enough. There are great examples where major, mature properties, that were well run by major operators, were perceived to be reaching the end of their economic life and they divested them to smaller companies.
It has been interesting to observe how these smaller companies have found producible reserves within those properties that the larger companies did not find. Obviously the majors could have found them. They have the technical expertise, but they chose not to pay attention. Was that because of being clueless, operating with bad data management, or having better opportunities elsewhere on the planet? I would argue in every case there is a bad data management component in the poor decision.
Look for part two in the next installment of Inside Synergy.
