To many, the heavy oil fields in the Bakersfield area of California might not seem like the first place a company would look to implement an intelligent development strategy. After all, these fields have been around for decades, they’ve been under steam flood for many years and they’re about as reliable as Old Faithful in terms of predictable production.

To Chevron, however, the sheer size of the reserves makes these fields perfect candidates

Figure 1. One of the “splitigators” at Kern River. Splitigators control and measure steam going into the field. (Photos by Rhonda Duey)
for a new way of thinking about asset management. The company is successfully using its “i-field” approach on several large heavy oil fields and has also established a program in conjunction with the University of Southern California to provide graduate training and research in a new discipline at the Center for Interactive Smart Oilfield Technologies (CiSOFT).
The i-field approach consists of
several initiatives to streamline field operations:
• Enabling faster and better asset management decision-making;
• Taking a multidiscipline, cross-workflow and systems view to asset management, not just “my world;”
• Making the right data easily available to whomever needs it; and
• Balancing automatic and human decision-making.

According to information on its Web site, Chevron is the largest producer of oil equivalent in California, with average net daily production of 202,000 bo, 101 MMcfg and 5,000 bbl of natural gas liquids in 2006. The company operates primarily in the San Joaquin Valley, where its three major crude oil fields — Kern River, Midway Sunset and Cymric — had combined net production of 155,000 boe/d in 2006.

With heavy oil making up 80% of the crude oil production, the company continues to focus on heat management in the recovery of these reserves. By increasing the number of wells being drilled, Chevron has reduced by nearly half the decline rate from these fields between 2004 and 2006.

The i-field approach
At the San Ardo field, another San Joaquin Valley heavy oil field, the project design focuses on automated optimized workflows resulting in better integration, enhanced decision-making and reliable field execution. Despite the age of these fields, Chevron considers new technology to be critical to i-field success because it enables integration, decision-making and execution.

At San Ardo, a multilevel approach was taken. First, team members identified critical outcomes for success. They also identified key work processes in heat, well and water management that needed to be highly reliable.

Connections were established between these work processes where integration and collaboration were critical to reliable field execution.

Human behaviors were identified which were driving the results, and a “trajectory and signpost” approach was developed to link short-term decisions with longer-term reservoir management tools. Then preferred alternatives for transforming key work processes were identified, including necessary hardware, instrumentation, decision support software, collaboration, change management and new technology requirements.

Finally, a phased execution plan was created to capture value in time for the major field development project. This included protecting intellectual property rights, integrating research and development efforts where technology gaps existed, and sharing lessons learned to accelerate change at other locations.

Overall, the i-field approach at San Ardo was intended to accomplish several specific objectives:
• Ensuring that the concurrent dewatering and steamflooding of the Lombardi reservoir tracks the project plan by creating an optimum plan to implement recommended latent heat requirements and optimize the steam system;
• Minimizing delays in the pattern-by-pattern field development by creating flexibility in response to reservoir and well changes and providing a collaboration environment to minimize down-time;
• Managing the complexity of the water management processes to ensure discharge requirements are met and providing decision support tools for responding to changing conditions; and
• Managing the interfaces between Chevron and its contractors in the field.
The project team improved 21 work processes, and these ideas are being used in other fields in the area, including the Kern River heavy oil field.

Kern River
Kern River was discovered 1899 and has been on steam flood for 50 years. Finding and producing the oil is not particularly challenging, but doing so in an optimal way has been an evolutionary process.

“Kern River is a great place to experiment,” said Greta Lydecker, general manager, asset
Figure 2. A Chevron engineer shows some of the valves that control steam injection into the reservoir.
development for Chevron’s San Joaquin Valley Strategic Business Unit. “We know the oil’s down there, but we have to find ways to produce it. We bring people to spend time here to work on development, and we leverage that as a key capability.”

The field is home to Chevron’s Heavy Oil Center and is another one of Chevron’s i-fields. The team is taking advantage of developments in sensors, monitoring and optimization tools that anticipate and plan based on what’s happening real-time and continually adjust to operating circumstances.

At Kern River, some wells are injectors, pumping steam into the ground. The steam lowers the viscosity of the oil and allows it to percolate through the formation to the well bore and then to the surface. Other wells are observation wells, instrumented with sophisticated sensors to track the steam movement.

Information from these sensors is routinely plugged into reservoir models that indicate how the field is performing. Kern River employees feel this gives them a better ability to make informed operational decisions. It may seem that the amount of data being collected from 660 observation wells and miles of core is overkill, but even minute improvements in reservoir management multiplied by the field’s 11,000 wells result in a major impact on the bottom line.
In addition to reservoir monitoring, Kern River is also an experiment in workflow optimization, with the goal being to allow the production engineers to “manage by exception.”

“Our current performance is like Excel with plug-ins,” said James Ouimette, consulting engineer for process planning. “To get past that, we had to identify work processes.

We identified 104 steps just to do steam injection. That’s a cumbersome process that thwarts the ability to target the field.”

The i-field approach helps to streamline that process and also helps the company capture knowledge from its older workers and raise the skill level of new employees. “Collaboration, work processes and demographics are our key drivers,” Ouimette said.

CiSOFT
Another form of knowledge transfer is taking place at CiSOFT. When Chevron officials went looking for an academic partner to help them research smart oilfield technologies, they had to ask themselves what strengths such a university would need. A petroleum engineering department was obvious, but so many other diverse technologies intersect at the smart field level that finding a school that could speak to all of them seemed a monumental task.

But the University of Southern California ended up fitting the bill perfectly. In addition to having one of the oldest petroleum engineering departments in the United States, the school, located in downtown Los Angeles, is also recognized for its work in information and communications technology and in advanced visualization. It also has strong relationships with the defense and even the entertainment industry, making it a unique match.

The resulting program, CiSOFT, was launched in 2003. Headed jointly by Iraj Ershaghi at USC’s Viterbi School of Engineering and Mike Hauser from Chevron, it currently has about 30 Ph.D. students in petroleum engineering, industrial engineering, information sciences and electrical engineering. It graduated its first students in May 2006.

Offered through the Distance Education Network of Viterbi School of Engineering, it is the first program of its kind providing graduate education in smart oilfield technologies.

Ershaghi was awarded the Society of Petroleum Engineers’ Education Service Award in 2006 for helping to establish the program. The award in part acknowledged the online availability of the program, enabling engineers in the heavy oil fields around Bakersfield to avoid the 4-hour round trip to Los Angeles to pursue their studies.

Collaboration between oil companies and universities is common but not always a smooth process. Ershaghi, though, said that so far the relationship is “a perfect match.” It has had a few growing pains — faculty from the computer science department had to have a crash course in petroleum engineering, for instance — but now, Ershaghi said, “we’ve been blessed with the right kind of chemistry of the individuals working together, both on Chevron and on our side. The focus emphasizes academics, making sure we have graduates coming up who are really making contributions, the future leaders of industry.”

The combined teams first met in February 2004 to discuss technology needs, and there were many. Realizing that personnel and funding would not allow them to research 30 or 40 different technologies, they identified seven major “challenge areas” that continue to drive the research. These include integrated asset management, well productivity improvement, robotics and artificial intelligence, embedded and networked systems, reservoir management, data management tools, and immersive visualization.

While these are broad and complicated topics, this is not blue-sky, long-term research. “We can’t wait for 5 years for someone else to develop solutions to these problems,” Ershaghi said. The center maintains a philosophy of thinking about things in new ways but also realizes that existing technology can be used in different applications to solve these problems. And even though the oil industry has a terrible reputation for the slowness of its technology uptake, he said that some of the ideas his team has come up with are already finding their way into the operation.

“It’s not like shooting in the dark, trying to come up with a solution for a problem,” he said. “It’s a matter of finding the really big gap areas.”

Hauser stated in a press release issued earlier this year that the first actual prototype uses of technology will be in the field before the end of 2007. Among the projects in development are initiatives to use enhanced analysis of real-time information about oilfield conditions, which he singled out as “the one most likely to quickly pay the largest benefits.”

Additionally, two new “grand challenges” were revealed — a new method for producing brownfields after “all the easy stuff has already been tried,” and the ability to design, in deep detail, exactly how to get the maximum production from a greenfield.