That formidable oilman’s user club, the Petrotechnical Open Software Corporation, founded four years ago to bludgeon suppliers into putting together open systems that did what they said they would, has popped up again to give us a view of its progress so far, describing it as a glimpse of true open systems. The first two […]
That formidable oilman’s user club, the Petrotechnical Open Software Corporation, founded four years ago to bludgeon suppliers into putting together open systems that did what they said they would, has popped up again to give us a view of its progress so far, describing it as a glimpse of true open systems. The first two phases of club’s Industrial Pilot Project, one run in the organisation’s head office in Houston, Texas, and the other in London, have been completed. According to the oilmen, they show that moving to common open standards can save up to 25% or 30% by making better use of scientists’ time. Oil companies such as Mobil Oil Co Inc reckon that their geophysicists spend up to 80% of their time simply finding and formatting data and then, in the words of Lance Johnson chairman of Mobil’s North Sea operations, Putting it back into the database for the next user. Improving that figure will mean a tremendous fillip for the bottom line of oil companies, so it is easy to see why the major oil companies, including British Petroleum Plc, Elf Aquitaine SA, Roayl Dutch Shell Plc, Texaco Inc and Chevron Inc, have been keen to back the Petrotechnicals.
Ferrets in a sack
The big prize, according to Johnson, is not just an incremental improvement in the way the oil companies do business, but a giant step forward. And no, this outbreak of harmony in search of common computer standards will not stop the notoriously competitive oil industry from fighting like ferrets in a sack in other areas, says Johnson. If the prize for the oil industry is clear, however, what is not so obvious is what the benefits are going to be for the computer industry. Since the oil industry spends a great deal on its computers – between $5,000m and $10,000m every year, the suppliers are keen to be to be seen to support the Corporation and its initiatives. They are not the only ones: of the group’s membership roster of almost 90 names, only 17 are oil companies. The rest are computer vendors, service companies in the oil and gas industry, and government and research bodies. So will the work, described by Larry Gahagan, the organisation’s Industrial Pilot Project director in Houston, as taking things that already exist and fitting them into the world of exploration and production, make a difference to vendors’ existing open systems strategies and products? There are signs that the biggest impact is going to be on some of the smaller, more specialised vendors involved in the project, rather than on the larger suppliers. Nonetheless, the demonstration by the Petrotechnicals, with both of Industrial Pilot Projects, that tying together various open systems models actually does work is an important step forward in its road map.
By Ray Hegerty
There are four parts to the Corporation view of open systems: base standards, the data model, data access and the user interface. The London-based part of the Industrial Pilot Project tested the practical viability of this approach in one of the main oil industry processes, prospect evaluation, where the risks and costs of drilling a particular area are weighed up. Using its Epicentre common database model, the project demonstrated that it would be possible to cut by up to a third the time taken to carry out such a process, according to project manager Gerard Huard. The details of both the first pilot phases are freely available in book form, and so far, according to Gahagan, 500 copies of the Houston project results have been sent out. The Corporation itself is acutely aware that it has to produce more than a couple of books to justify the investment in both staff and money being made by its members on both sides of the vendor-customer fence. Mobil, for instance, is ploughing subscriptions totalling $1.2m a year into the group in the hope that something concrete will emerge. It is, acknowledged Johnson, an act of faith, but he says that Mobil is expecting significant results from its investment. And what if, at the end of all the effort, it does not deliver? Johnson said the whole project would be backed as
long as it was making progess and holding out the promise of business benefits. If it stops looking that way, however, he made it clear that a couple of years down the line the industry would have little hesitation in pulling the plug, so the Petrotechnicals know they still have a lot to prove. One of the major beneficiaries of the latest Petrotechnical Open Software Corp initiative is UniSQL Inc of Austin, Texas. UniSQL, the object-oriented relational database outfit, funded by Nippon Telegraph and Telephone Corp, has successfully positioned itself as the central data store for the Industry Pilot Project. The financial rewards will hardly set UniSQL’s accountants and treasury department jumping for joy, but the prospect of a market which spends anything up to $10,000m annually for computer systems and services is not one to be ignored.
The Petrotechnical criteria are based on the need to reduce the cost and improve the efficiency of the exploration and production industry. With the majority of the Corporation’s members and affiliates consisting of vendors, the Industrial Pilot Project is an ambitious attempt to validate the standards of the group. Larry Gahagan, project manager and system architect for BP Exploration, places the future success of the pilot squarely on the shoulders of vendors willing to meet us more than half way. In practice, companies such as UniSQL have devoted much time, effort and no doubt resources, in attempting to woo the oil companies. If their investment pays off, they are hoping for a 5% to 10% leverage in the overall market. Larger companies such as Oracle Corp are more resistant to the possibilities that the Corporation really will fulfil its ambitious aims. As a result, it is hedging its bets, and providing Oracle systems with UniSQL’s middleware.