A survey of engineers revealed problems with production software and future needs for integrated, Web-based programs.

Engineers are covering more ground these days. Individual engineers are responsible for more fields and more wells than ever before (Figure 1). In recent years oil companies have shed 4% of their engineers through mergers and downsizing (Figure 2). The result? Even companies that were lean in the past now spread their expert resources thinner than ever - and are actively seeking tools and strategies to help engineers work smarter and faster. It's about time engineers begin to reap the productivity benefits of integrated software systems.
Hart's E&P surveyed 15 production and reservoir engineers in large and small oil companies around the world about software for the production environment. Unless they're building a custom solution - and a few are considering that - most say the software vendors don't yet have what they need. But they had a lot to say about what that system should look like.
Data: A major issue
First, data integration is a major issue. The top production-related priorities in the companies surveyed are:
• shorten the time to first oil on new developments;
• optimize production from existing fields; and
• improve data access and integration.
One reservoir engineer at a major oil company said, "I think the most significant challenge we face is data. Systems have improved, and we're getting a lot more data thrown at us. And a lot more data can be digitized and moved into this environment. The problem is that it's coming from dissimilar sources, so being able to integrate and make sense of it is the problem." Another way to look at the data problem, he said, is that the "data has value for a longer time than we're likely to focus on the asset."
Chris Stevenson of Forest Oil said, "One of the big things we're working on is tracking our production, trying to set up all automated systems to pull the production data in from the field and get it up to the production office and do our production accounting and get it back out to our business units." The systems for gathering, formatting and moving data are critical and involve significant technology issues. Stevenson stressed the importance of accurate production reporting for keeping Wall Street informed. "It's important to understand immediately why goals are not met. Production really drives your company."
Real-time data
Real-time data is generated in the industry. In the majority of the companies surveyed, engineers got daily production reports in their area of responsibility, and many reported some access to real-time data. However, most acknowledge their access to current data is spotty at best. "I get some things like pressures from the downhole gauge every 10 seconds on an internal Web site. In other areas I'm better off to consult public records before I get the company stuff," one engineer said.
From listening to engineers describe how production data are collected, it became clear that the ideal production optimization system would require an integrated approach to data capture. For example, it was common for engineers from large companies to say that somewhere in the company they probably used every method known to the industry to capture data from producing wells. From field operators who travel to each well, write down production figures in a tally book and call or fax the report in to a central location, to the most sophisticated surveillance system that not only captures the data automatically but allows remote intervention, they use it all.
The problem is not so much that the systems are inefficient, but that they are different. The data goes into different databases, in different locations and in different formats. That makes putting together production reports for a business unit or company a matter of getting the data from multiple systems and integrating them by hand, a time-consuming process.
Even large companies don't have fully automated production data-capture systems. One engineer estimated that in her business unit only 10% of the wells were on SCADA systems. Putting the real-time data question in context, Augustin Diz of Repsol/YPF observed, "Real-time data only makes sense if you can really automate it. In our experience it's too costly in many areas: for example, in an area with 8,000 wells producing low volumes."
The benefits of real-time production data are universally recognized. "It allows you to act more quickly when you see a well going down and to see trends you can act on more quickly," said a reservoir engineer from a major oil company. Said another, "I'm going to be continually looking to optimize my performance, and if my data is a month or two out of date - as it usually is - I'm not going to be that efficient. But if I can detect a production hiccup immediately, I can maybe diagnose a problem with the casing immediately and not a month from now."
Engineers also see a downside to real-time data. An engineer from a small independent firm isn't too keen on real-time data because it contributes to data overload. "A prime example is when you bring a well on after completion and you're looking at the flow-back on an hourly basis. You get caught up in the minutia and forget to step back and look at the big picture."
Another engineer lamented that if he were doing his job properly he'd look at every well every day. "But I can't. Don't have the time. But if I had a computerized system that goes in and reviews and gives me a report..." he dreamed. "It's a big challenge just looking at all the data," said another engineer, "because if you looked at it, it could tell you things, like which well needs a workover - give you insights like that."
Another response to data overload, according to some engineers, is to look for more ways to visualize data. "I'm a big believer in visualization, but the key thing isn't just visualizing it but making it interactive. I want to see what happens when I change something. We've got some pretty high-powered visualization systems now and we see Space Center-type technology that will soon bring more visual benefits."
Rollup and drilldown
More and more, it seems, companies want individual employees to understand how their work contributes to the company's success. Being able to roll up and drill down to track key performance measures like production, reserves, costs, revenues and income not only helps the accountants and strategic planners understand what makes the company tick, but it also lets business units and asset teams - even individuals - see their contribution to that 15% production increase.
"Everyone in the company has a performance management contract with his or her supervisor," said Trem Smith, director of reservoir management and technology for Chevron's Thailand business unit. "There are employees on the front lines today who actually understand their contribution to barrels (of oil). So the rolling up of the key metrics is essential."
Others looked at rollup and drilldown from a portfolio management point of view. "I need to be able to drill down and say, from the perspective of my business model, how do I compare these assets and make decisions about trading them off? That's a big activity with independents - balancing assets. If I see less value in some asset I want to trade it to someone whose business model is different and who maybe sees more value in that asset."
For many, a key feature of any production optimization system is the link to the reserves reporting database. One engineer said connecting the production database to the reserves database was the principal value of integrating different software packages.
Web-enabled: 'The way to go'
More than one engineer already has a vision of how the ideal software system for engineers would work. "Ideal would be if the production or reservoir engineer could come in, open up a browser and take a look at what's happened in the last 24 hours in the wells or reservoirs or leases under his responsibility."
"Yes, Web-enabled is the way to go," said another.
No one was surprised when the suggestion was made that the next generation of production optimization software would be a fully Web-enabled PC system. Those companies without Web access to critical data have made a decision to go in that direction.
Platform independence not only means those on Unix and PC systems could use the same system, but that those traveling with laptops, personal digital assistants or even cell phones could access critical data. One manager said he wants to be able to "communicate with someone who's traveling and has only a Web appliance. I'd like to be able to talk to that person about an optimization issue so that person is actually in the dialog and I'm using their brain power." Another response, only partly in jest, was that you could "go to an Internet cafe on the beach and check the day's production."
Equally as attractive is the integration possible on the Web. Asked about the benefits of connectivity among applications that were previously separate, one engineer rated it 9,000 in importance on a scale of 0 to 10. Another suggested that a Web-enabled system pulling data from different systems could "go beyond production data gathering and show you costs on the other side of the page."
Intelligent optimization
Several respondents said their focus for the future would be on "intelligent wells." Such wells could make their own decisions about optimizing gas lifts or doing a downhole separation, for example. "We need to move to a smart well manager," one respondent said. One engineering manager described how his company was planning to make production more efficient: "Layered on top of exotic surveillance systems that allow you to exercise control from a distance would be the so-called expert systems which would automatically intervene to optimize rates."
Engineers were asked about their interest in systems that could help them decide whether to infill drill or workover to increase production. A few were downright skeptical, but most said they like the ideal of an intelligent system, provided there was enough flexibility to adapt to different situations and accommodate different company workflows.
The difficulties of integration
Participants in the study mentioned the following data integration problems:
• different methods of gathering field data;
• multiple instances of databases with no synchronization;
• local databases that are not rolled up;
• public data with data formats that don't match internal data conventions;
• weak communications infrastructure environments;
• no methodology to link production history to simulators, economics, revenue accounting or regulatory reporting; and
• no links to reservoir models, drilling and completion data, or economics and portfolio management systems.
Because the data integration problem is so significant, most engineers believed the cost of a new integrated software system would be only a fraction of the total cost of solving the data access and integration problem. Chevron's Smith cautioned that companies will have to move deliberately to change data integration and work processes because, "When you talk about behavior change, we've discovered the hard way that becoming Web-enabled requires people to change the way they do business. And if you underestimate the time and the cost of getting that done, you'll fail."

Comments