News Article 08/08/2006

Structuring oil and gas information

Digital Energy Journal, Feature Articles, Aug 08, 2006

There has been plenty of talk in the oil and gas industry about the problems of having too much unstructured information.
Information is not provided to the right people at the right time; people do not learn from the experiences of other people doing the same task elsewhere in the company; and it is very hard, or maybe even impossible, to find out why certain decisions were made in the past.
Most people agree that efficiencies and safety improvements could be made if people could co-ordinate their work better.
The challenge is how to set about making a software program that can co-ordinate people's activities and store the relevant information as people go along, without creating additional work for people.
UK software company Ulysses Systems has got close to creating a solution for the commercial maritime industry, where its software is being used onboard around 1,000 oil tankers, offshore support vessels and other vessels; it is now seeking business opportunities in the oil and gas industry. 50 shipping companies are using the software, twice as many customers as it had 18 months ago.
It recently signed up French marine services giant Bourbon Offshore, which is using the software onboard its 130 vessels, including 7 multi-purpose supply vessels, 21 anchor handling tug supply vessels, 22 platform supply vessels, 3 terminal tugs, 8 fast support and intervention vessels, and 11 crew boats.
Bourbon purchased the software, Ulysses says, as a means of co-ordinating and consolidating all of its corporate communications plus maintenance, purchasing, crewing and document co-ordination, and because it thought it was the easiest to use.
The information is structured around company activities, as defined by the Bourbon employees.

Ulysses' solution
Ulysses solution is about designing ''activity models'', creating software which can be used to co-ordinate regularly completed tasks and store the information about them, so it can be accessed immediately in other parts of the company, and kept as a record for the future.
Ulysses does not try to drag people away from their spreadsheets and e-mails; but all emails and spreadsheets are automatically indexed so that they can easily be accessed at a later date.
Activity models can be built for any task people commonly refer to; the tasks can be relatively predictable, like a planned maintenance task, or relatively complex and as yet unpredictable, like bringing in experts around the world to try to explain a drop in production.
To date, Ulysses has built software modules to manage tasks and information flow related to maintenance, purchasing, crew management and documentation.
The software system is not so rigid that it requires that staff standardise their business processes; they just need common reference points.
The software is possibly best suited for business processes which are prone to failure or unexpected success, because they can easily record the reasons failures happened or successes happened, so the organisation can learn from them.
So, for example, if a company wants to find out why boiler controls were changed 5 years ago, all the communications and documents related to this decision will be automatically stored and can be retrieved by their relationship to current concerns.
If there is an occurrence the software system can retrieve structured information about everything, which was relevant to the occurrence as well, as how it was remedied.
This requires no special reporting or indexing by users so information collects in the system from which subsequent users can learn so as to avoid the same occurrence.
The software can provide information to users, taken from other parts of the system, about what they are currently doing.
The relationships between activities can change continuously and the system can be set up to adapt to the changes.
If staff members like to use e-mails and spreadsheets, they are automatically indexed.
When individuals log onto their computers every day, they see a list of tasks outstanding, which need their attention. When they click on them, the software provides ready information they need.
The software can also provide rapid access to information, which might be needed when specific tasks are being performed.
Ulysses stresses that this is not calling its software "knowledge management" software, which operates in a very different way, normally requiring people to type in their experiences into the system, rather than indexing the information they create anyway.
Ulysses believes that activity co-ordination software such as this is most suitable for work environments which have a small group of employees doing many different tasks but doing the same processes repeatedly (e.g. corporate communications co-ordination and purchasing, maintenance, crew management), where there is critical issues which justify the software investment (e.g. high risk / high liability, expensive assets, regulatory requirements.

Here are some specific examples of how activity co-ordination software can work.
One offshore vessel master told of an oil rig captain who asked the vessel to accept drums of chemicals, but refused to disclose what the specific chemical was, on the grounds that the vessel operator had signed a contract that it would accept all deliveries from the rig.
The vessel master needed rapid access to company policy information, to tell him what to do in such a circumstance, whether to accept the drums or not, which the Ulysses Systems software could make available.
Another example is purchasing. If someone needs to make a purchase, the software can be configured to have information readily available about the last time similar items were bought in the company, if the items were actually used (or are sitting in storage somewhere), who the supplier was, what the price was, and how well the product performed.
Another example, taken from the maritime industry, is of a ship with two ballast systems, one of which is suddenly not operable and the other with a hidden defect that could only become relevant under certain circumstance.
The engineer knows that he can pump the ballast in twice as much time (half speed), so the vessel does not need to be taken out of service.
However this means that the second pump needs to pump and also to purge the tanks at full capacity, which it may well not be able to do and while this may have been its condition for a long time it now becomes highly relevant to the safety of the vessel.
If ballast is being pumped out and cargo is being loaded simultaneously, the minute by minute loading characteristics of the vessel will change with just one ballast pump, which could lead to the vessel being overloaded at certain points or dangerously strained or touching bottom during loading.
A structured software system such as Ulysses can relate the process of planning the loading with the process of running all the related machinery and can highlight potential hidden problems in the machinery that may not come to notice until its too late.
A further example of typical problems with e-mail systems is a document, which belongs in three different folders because it is related to three different issues. Using a Microsoft Office type system, the user has to save it three times, and then might find that one version is being updated and the other isn't, causing all kinds of problems.
With a more structured software system, the document is only saved once, but the system is set up so people can assess it when they need it for the three different issues it relates to.
It might be possible to access all of this information by searching through company intranets, but it would take a lot longer, filtering through lots of poor matches; there is also the problem that staff might not know what they are looking for because they are not aware of its existence.
For example, in the case above with the ballast pump, the chief officer would not necessarily search for documents to find out if the second working pump had a defect which prevented it from being able to pump tanks at full capacity this needs a system to ''push'' the information rather than wait for someone to go out and find it.

Text search vs structured information
Most people are comfortable doing text-based searches, both on their own PCs and on Internet search engines, and have a good knowledge about the type of information that can be retrieved easily with this kind of search.
Most of us are even very reliant on text based searches to find all the information we think we need for example, searching for someone's phone number on the internet using Google, rather than writing it in our phone directory.
What is not so readily understood is how much a structured data system can help provide the right information when people need it?
On a system like the one Ulysses develops, since all information created during work is automatically indexed as it is created; there is no need to file or label anything, and there is no vital information on people's personal PCs, or lost in their e-mail systems.
"Outlook is not an information management system," says Mr Lyras. "It is a linear messaging system. It is not even a shared document system.
"Even shared document systems on the market have no model of what people discuss and do in the company. "So the system cannot help manage its own content within the pattern of the organisation.
Ours is a whole different way of looking at how you manage the information. It's a revolution in how you hold information. That's much closer to how people classify information in their own mind."