Logo GISdevelopment.net

GISdevelopment > Proceedings > GITA > 2000


GITA 2002 | GITA 2001 | GITA 2000 | GITA 1999 | GITA 1998 | GITA 1997 |  
Sessions

Data development and evolution

Engineering and design applications

Exploiting field and mobile technologies

Invited presentations

It's a brave new world

Leveraging web-based technologies

Mobilizing the enterprise

Operations support

People issues

System architecture

The best of the rest

Uniting the enterprise

User perspectives

Work management solutions



GITA 2000


Uniting The Enterprise


The virtual pipeline: Operations insight fuels marketing decisions


Marketers also have sophisticated information systems, which let them solicit and manage nominations, and make decisions about which to confirm, reject, or curtail. This is roughly analogous to the problem airline agents face: multiple agents are selling seats on many airplanes at many prices. Each agent needs to know which seats are available at each price, and with each sale the inventory and prices change. Like airline agents, pipeline marketers have little control over prices, which are established in long-term contracts. But unlike airline agents, pipeline marketers have little information about their inventory; typically, they have a weekly spreadsheet showing how much gas can flow between each section of their pipeline. This "nominal capacity" is their only regular source of capability information. As a result, pipelines are often undersubscribed, costing the company revenue, or oversubscribed, upsetting customers or incurring penalties.

In a deregulated spot market, pipeline marketers will operate less like airline agents and more like airline ticket auctioneers. To succeed, pipeline marketers need to know the actual available capacity of their pipeline and the marginal cost of transportation for each nomination. They also need to know what the market is doing and will do over the next few days. Furthermore, they need to access this information in real time, share it among all other marketers, and update the database with each decision made. Each needs to know which seats are available. Unlike the airlines, pipeline marketers also need to know their marginal cost of filling each seat and what price the market will bear.

The only way to achieve this dream of informed, and hopefully profitable decision making, is to integrate operational and marketing information and publish it throughout the business, in accessible, current, and appropriate formats. Our research into this problem has led us to a complex document we call a "virtual pipeline," which organizes both marketing and operations information to be flexibly useful to many people in the company.

A virtual pipeline
A virtual pipeline is a set of informational displays - maps, charts, reports - that are broadly accessible, and provide operational and marketing insight to support business decision-making. These displays are built from information gathered from various technical and business information systems, which is consolidated into a standard set of business objects, and delivered on demand as maps, charts, and reports via intranet or Internet to marketers, operators, and executives.

Through work with pipeline industry clients, we have established design criteria for several aspects of virtual pipelines. The following sections present the technical architecture, information requirements, presentation requirements, and organizational impact of a virtual pipeline.

Technical Architecture
The technical architecture for a virtual pipeline consists of subsystems for information generation, information transmission, and information distribution.

Information is generated from operations and marketing systems, including nominations and gas management (product accounting) systems, SCADA systems, and hydraulic simulations of pipeline operations. Optimizing and predictive simulations provide information about future states of the pipeline. Forecasters are also used to predict supply and demand. Critical real-time data comes from Internet-based feeds of weather and market pricing data.

Information is transmitted through a set of system interfaces that harvest information from many sources and consolidate it into a standard set of business objects, representing all the important constructs in the virtual pipeline. These objects range from tangible operations assets such as compressor stations, to more abstract marketing objects such as contract pools. For these objects to deliver value they must be comparable and available. This implies a logical data model that is consistent among objects, especially with regard to names and units of measure. It also demands a standard corporate communications layer, including a standard object request broker. Our experience suggests that an open standard technology such as CORBA/IDL is most robust and maintainable in the corporate environment.

Information is distributed as web-based displays, created on demand, to present highly processed information in compelling, graphical images. A virtual pipeline requires many graphical tools for rendering appropriate displays, depending on information content and user preferences. Much valuable information about a pipeline is dynamic, which requires the information distribution technology to render displays on the server but subsequently provide active updates to an interactive browser-based client. Technologies critical to data distribution include geographic information systems (GIS) for map-oriented display and interaction, ActiveX or equivalent controls for schematic displays, and graphing, charting and reporting components for other displays.

Page 2 of 3
| Previous | Next |

Applications | Technology | Policy | History | News | Tenders | Events | Interviews | Career | Companies | Country Pages | Books | Publications | Education | Glossary | Tutorials | Downloads | Site Map | Subscribe | GIS@development Magazine | Updates | Guest Book