Using Geographic Information Systems in the Petroleum Industry
William N. Wally
Chevron Petroleum Technology Company
Houston, Texas
| Abstract Of The Paper & The Profile of The Speaker | Speaker Index | Paper Title Index |
Background - GIS at Chevron
In 1987, the first copy of the Oracle Relational Database Management System (RDBMS) at Chevron was installed at Chevron Oil Field Research Company (COFRC), to evaluate whether such technology could be used to support seismic and well data applications running on our engineering and geological workstations. The benefits of having all the data available in a single shared database were apparent, and Oracle has since become a defacto standard. This technology laid the groundwork for building databases that later could be easily linked to a GIS.
In 1990, several of us attended "Operation GIS", presented at the National Computer Graphics Conference in Los Angeles. "Operation GIS" was an effort to apply "off-the-shelf" vendor software to typical oil industry problems (e.g. "find all the wells in company-owned leases that expire before the end of the year which are less than a mile from any pipeline"). Though at that time no single product could solve all the "problems", the GIS vendors came the closest, and most of the "traditional" oil-industry service companies declined to even participate.
In 1991, Chevron purchased its first copy of ARC/Info. Unfortunately, at that time there were deficiencies that, for us, really prevented serious use. Missing were "regions" (overlapping and non-contiguous polygons - critical for land/lease management) , "dynamic segmentation" (critical for seismic line annotation), and a full function PC product (since PCs greatly outnumber other computers).
Accordingly, Chevron proceeded with development of an in-house PC-based Land Information System, and the UNIX-based Graphical Element Editor (GEE), both of which were deployed in 1992. However, it was clear that GIS technology had promise. And we saw that ESRI in particular was really taking to heart the criticisms and problems voiced by us oil companies. Accordingly, we continued to closely follow GIS technology developments, especially at ESRI.
In 1994, ESRI's introduction of ArcView (which runs on PCs and Macs as well as UNIX) and new improvements to ARC/Info finally made the difference. And, we appreciated how the ESRI Petroleum User Group (PUG) -- Chevron was one of the oil companies on the PUG steering committee -- was proving to be an effective forum to convey oil industry needs to ESRI senior management, who responded by making important product improvements clearly with our requirements in mind.
Since 1995 Chevron Overseas Petroleum has had several initiatives to apply GIS technology on an in-country enterprise-wide basis. We are convinced that GIS technology is seriously relevant to our industry, offering unique, affordable, comprehensive solutions, readily supporting entities including:
wells
pipelines
seismic locations
facilities
engineering drawings
photographs of facilities, wellheads, etc.
safety and environmental reports
spills
land ownership and permits
roads
rivers
villages
digital orthophotos
satellite images
Several GIS initiatives are well on the way to becoming production systems integrated into our business processes, with support groups and an increasing number of end-users who continue to discover new ways of benefiting from this technology.
Drivers to use GIS technology in the Petroleum Industry:
GIS technology has increased its relevance to the petroleum industry because of two driving forces:
technical
business
Technical drivers:
Hardware : Fast and relatively inexpensive desktop computers, with high-quality graphics monitors, connected to reliable networks, are becoming fairly common in our business. For a few thousand dollars, one can now buy large-format color plotters that used to be an order of magnitude more expensive as well as slower.
Thus one of the primary information-bearing vehicles of our business &endash; the map &endash; can now be created "on demand", easily, quickly, and accurately using GIS technology.
Software : GIS software like ArcView is much easier to use than ever before. The fact that it runs on PCs and Macs makes it available to everyone in the oil industry operation, including for example:
- management reviewing reservoir financial performance and competitor activity
- engineers siting well surface locations, who need to know where roads and pipelines are, as well as lease boundaries, population centers, environmentally sensitive areas, etc.
- contractors permitting access routes or handling damage claims
- emergency response teams
Data: Until recently, GIS technology was handicapped by high cost and long delays building geographic databases. In the past few years, there have been major improvements in the availability, accuracy, and cost of spatial data. For example:
- high-precision global positioning systems (GPS) demonstrate that many existing maps are inaccurate and must be redone.
- high-quality GIS datasets from government and non-profit agencies are now easily retrieved from the Internet
- high-resolution satellite images (especially the recently-announced 1-meter data) produce very much larger image files (e.g. 1000 times larger), and will result in derived vector files larger still. Advanced computer systems are needed to handle such large datasets, and ESRI's Spatial Database Engine (SDE) is an example of powerful technology essential to properly manage this kind of data.
While not all such data is accurate enough for oil industry use, the quality is continually improving. It seems reasonable to expect much higher resolution data to be commonplace in a few years time, and to consider reengineering business processes affected by such data availability.
Relevant functionality : ESRI deserves much credit for forming an oil industry vertical marketing organization, staffed by former oil industry professionals. Their products have significantly evolved in the past 3 years. New features like "regions" (overlapping/disjoint polygons), "dynamic segmentation", and overposting elimination (based on work done by Amoco), have eliminated most of the shortcomings that previously handicapped use by the oil industry.
Business drivers to use GIS technology:
Multidisciplinary asset teams : It is now not unusual for asset teams consisting of disciplines such as geophysics, geology, petroleum engineering, facilities engineering, land/legal, and safety and environment, to all cooperate managing a reservoir. Such groups clearly benefit from shared databases, especially for map data showing locations of key features like wells, pipelines, population centers, environmentally sensitive regions, etc.
Although traditionally some of these groups knew next to nothing about each others' work, they all share one common data "key" &endash; the location of the asset. Thus GIS technology, which of course also uses geographic location as its primary means of associating data and attributes, is well-suited to support such requirements.
High-cost of internally developed, proprietary software: Until recently, most oil companies used specially-designed computer software, often developed internally, for most scientific and technical data processing. This was because it was usually not possible to buy already-written software that did the job. The cost of such an approach can be great.
For example, it might cost an oil company $1,000,000 to develop a particular computer program. If it uses it at one site (e.g. the headquarters office), it will of course cost $1,000,000. And to deploy it throughout the corporation may be difficult since software support, education, and other infrastructure may not exist.
Leveraging the IT marketplace- I Industry-specific solutions
One alternative is to buy software from oil-industry service companies. These vendors can afford to spend more money developing the product in the first place, since they are selling to more than one customer. So, possibly their product might cost $10,000,000 to develop, but since they might expect to sell it to 100 oil companies, the unit cost works out to $100,000 per company. Thus a program that cost 10 times as much to develop can be bought for 1/10 the price!
3D seismic interpretation systems, well log processing and formation evaluation systems, and seismic signal processing systems are nearly always purchased from such industry-specific companies because of such economics, support and training, and quality.
But although these systems are much cheaper than in-house ones, they are still too expensive to place on every professional's desktop.
Leveraging the IT marketplace- II General-purpose solutions
If an oil company can purchase general-purpose software developed for the entire computer-using community, costs drop even more dramatically. So if a general-purpose product might cost $50,000,000 to develop but then is sold to 10,000 companies, the unit cost drops to $5,000. An example of this is the GIS program ArcView, which has taken over 500 man-years to develop yet sells for less than $2000. And this cost makes it reasonable to contemplate its widespread deployment, indeed on the desk of anyone who might benefit.
This figure depicts Chevron's planned GIS architecture, for a central-site operating company environment. Using modern relational database technology, all the data including seismic, wells, production, land, safety and environmental, and cultural/spatial appear to be in one place. That is, the end-user sees a "single database image".
However, the ArcView "shape" files remain for communication with remote computers. In addition to the "office" GIS environment (served by this SDE/Oracle/ARC system), we anticipate a "field" GIS environment, running on a laptop computer. For example, a permit man, or a surveyor scouting a new well location, might have a "snapshot" of the GIS database on his laptop computer. These "shape" files are not expected to serve as a GIS database. Rather they are working copies used to transport GIS data between the office and field environments.
Note also that because ARC/Info and ArcView are sharing the same data with GEE (the Chevron proprietary GIS) and other 3rd party software, we at Chevron Petroleum Technology Company (CPTC) expect to "offload" some of the more generic (and difficult to implement) GIS functionality to them. We then can concentrate on those activities where we truly add value, either because they are too oil-industry specific (or Chevron-specific) to reasonably expect ESRI or other third-party developers to bother with, or, perhaps, because we may have a proprietary advantage.
Future expectations
Although GIS technology brings lots of benefits to the oil industry already, it can still be made better. Where our requested improvements can be stated and implemented in a generic way, they can benefit all users of GIS. Some things on our "oil industry wish list" include:
Improved integration with our relational databases. Fortunately, the newly-announced Spatial Database Engine (SDE) makes a major leap in the right direction. However, we still have a way to go to link hardcopy and CAD data (though a large part of this problem results from poor record keeping and data management, which can't be fixed by the GIS...)
A truly global paradigm, where locations on the earth include "metadata" like Geodetic datum, Cartesian projection/spheroid parameters, etc. I am pleased to state that the next releases of ArcView and SDE appear to be comprehensively addressing this issue.
An even shorter learning curve to effectively be able to use GIS technology
Better management and access of seismic data and well logs
Improved "conflation" tools. "Conflation" is the process of merging two or more GIS datasets, so that the output has the highest-accuracy data from all the inputs. As surveying tools and GIS data from satellite and orthophoto images continues to improve, existing maps and GIS datasets must be "high-graded", i.e. adjusted to remain consistent with the newer data. Managing this process may well be our biggest challenge.
3-dimensional GIS. Right now, our use of GIS stops at the Earth's surface. To visualize subsurface reservoirs, we must change to completely different systems, which rarely present a "seamless" interface to the GIS.
Summary - GIS technology brings benefits to the oil industry
GIS technology and related hardware and software have advanced to the stage where they offer tangible technical and economic benefits to the petroleum industry. Not only do they improve our current business processes by furthering better data sharing and more accurate mapping, they also support efforts at "business process re-engineering", where our technical professionals redefine their activities as they are able to access critical data in new ways.
We are especially enthusiastic about the possibilities of SDE. SDE eliminates the need for proprietary files and access software, leverages the large GIS technology market, and offers higher performance than anything else of which we are aware. It enables much better organization of GIS data into central repositories, minimizing multiple copies of the same data (or, worse, almost the same data), and offers at last the possibility of different vendors application software accessing the same data.
For the oil industry, SDE could become a standard GIS access method if:
a common data model were adopted for standard datatypes like well and seismic locations
oil and service company take-up occurs
Chevron has been meeting with other oil and service companies, and using forums like this GIS conference, to see if others share our interest in creating and benefiting from such a GIS standard. The response so far has been encouraging.
| Abstract Of The Paper & The Profile of The Speaker | Speaker Index | Paper Title Index |
|
|
|
| |
