Moving at the speed of data
Tony Edwards of StepChange Global discusses how oil and gas operators are managing the use of data in operations, and how field workers and the control room are learning to trust each other more in this new digital age.
Capturing data in the oilfield is just one challenge for the age of the Digital Oilfield. Another significant issue is the culture change needed to bring all parts of the operation together to deliver on the potential value of all that data. In the second of a two-part series, Oil & Gas Engineering spoke with Tony Edwards, CEO of StepChange Global, a U.K.-based digital oilfield advisory and consultancy company focused on designing and integrating operation services, about the challenges in true integration of oilfield operations from a technology and a human perspective. Here are edited transcripts of that conversation:
OGE: In the last issue, we talked a lot about some of the technology enablers, we talked about some the cultural things and we’ll dip more into that, but why don’t we talk about how long integrated operations have been going on. Give us a brief history.
Edwards: There are certain elements of integrated operations, like remote control, that have been common throughout a considerable period of time. There were some people looking into using drilling data, certainly in the mid- to late 1990s and I was involved in a virtual teamwork project inside BP in the mid-90s that was looking at using desktop video compensate. But really integrated operations in digital oil field, as we recognize it today, really didn’t start going until around the 2002-2003 time period.
It was kicked off by a few companies that had access to some high bandwidth communications. There were a couple of operations in the Norwegian continental shelf where there were some fiber optics being laid between across the North Sea between Norway and the UK. At the time, the technology required some repeater stations. So these telecoms companies went to a couple of the operators there and said, ‘We’d like to put repeater stations on your platforms and we’ll give you 10 megabytes of bandwidth’, which in around 2002 and 2003 was an enormous amount of bandwidth to get back.
Once you got that bandwidth, you had the opportunity to bring the data from the offshore platforms, and that’s when some of the key pioneers started thinking about this and looking at this way of working.
And so a number of companies kicked off their digital oil field or integrated operations global programs around that time frame. The early pioneers were looking at down hole data and down hole sensor technology, and also looking at alternative completion geometries-gravitated wells, safe wells, and other things.
Once people in essential teams started seeing there was an advantage, these global programs began to spring out which were used to coordinate and take the best practice of learning and try it, encourage, and roll out this new way of working to other assets across the company.
Shell, BP, and Chevron are still rolling it out to assets, because whether you’ve got a company with a portfolio of operations which may be 30, 40, or even 50 operational sites, you know that takes some time, and within that portfolio of operations you’ve got some early adopters and you’ve got some people who want to jump on later on.
Very early on, everybody was literally making it up as they went along, and certainly there was an awful lot of collaboration between the big companies that was quite unusual. So we had regular meetings with Statoil and Chevron and Shell in the early days. Of course, once everybody worked out there might be competitive advantage in this, well you know the shutters came back down really quickly a few years later. But, when you were doing this for the first time, or the second time, or the third time, where did you go to learn how to do this?
What you still see to a certain extent in places like Norway is that the oil community between Statoil, Total, Conoco-Phillips, and BP all know each other and they’re all on each other’s doorsteps and actually they’ve learned a lot, almost as much off each other as they have off their own internal organizations elsewhere in the globe. But of course, once it became evident there was potentially intellectual property or certainly knowledge and knowhow and expertise in doing this, certainly, the normal kind of oil and gas behavior kicked in and people tended to rely on their internal programs once they got to point planning.
OGE: They kept their cards close to their chest?
Edwards: Yes. And what everybody did realize pretty early on was that it was not about technology; it’s about changing the way you work. For people very early on, there was kind of a ‘if we build it, they will come’ kind of approach. So we built a lot of IT systems, and we built some rooms, and of course they didn’t come along at all; they just carried on the old way. So then we realized that this really shouldn’t really regarded as a technology project at all; it should be regarded as transformation program which is technology enabled. Technology enables you in a different way, but actually, the hard work and hard yards is in the people, process, and organizational change. And I think that’s still true today, and I think that’s really changed.
The technology’s got to be good and it’s got to work well. And you can have it in such a way that it’s now more than three clicks of a mouse to get to any data, but just because you got that doesn’t mean to say you get the value. But when we did the first Intelligent Energy conference in Amsterdam in 2006, what was interesting there was there was a number of sessions just on people turning to process change and that’s been a common thread ever since. You can buy all of the technology and get all of the technology, but the technical bit is getting people to use it.
OGE: How are these teams formed that oversee the integration?
Edwards: You can have production engineers, and you can have facilities engineers, and you can have electrical engineers, or automation engineers. And you can think of it in terms of a functional or discipline view of the world. What we see is that if you think of it in what you might term a process view, what you see there is to optimize all of your gas and oil from one to the other is inherently multi-discipline. It needs reservoir civil engineers, it needs well engineers, it needs production engineers, it needs operations, facilities, pipeline personnel.
What we’re going to do is allow these multi-disciplinary teams, which are looking at their own data, but they’re located in a room supporting how do you want optimize delivery of maintenance or production, or inspection. It’s done from a multi-discipline point of view inside an asset. At the end of the day, an asset team for an operational offshore operational platform, for instance, is looking at his responsibility of running that asset in the best possible way.
If you’ve got real time data and information, it’s just actually just like having people walking around the building. The idea is to put all of the people in this collaborative environment, which in essence is a meeting group, and connect them there permanently all of the time with all of their data and information. The kind of success you get from them having access to that data and information and the is different than if you have a functional organization.
OGE: So you’ve kind of got this real time view of the data. But also you have some data that can be looked at for historical purposes.
Edwards: Yeah, it’s spanning data and it’s real-time data. When you get more advanced in this, you can put alerts in your data system and your control system because there’s a difference between alarms and alerts. Alarms are a proper controlling function-you obviously have a high-level trip. Alerts don’t go back into the control system, but it does alert an engineer that something is not quite right or something’s trending in the wrong way. And this is where you get to management by exception. The engineers have the data and information. And then they have to start observing it, and trending it, and then they start putting their own alerts in place to keep, and then they start building scripts to have this management process exception approach.
You know, what does a foreman has to do every morning when he gets in? He has to open every spreadsheet and look at the data and see if a well is OK or not. If you’ve got a lot of wells, that can take hours. Now what we’re doing is automating that, putting that on a screen on a wall, we’re actually saying, actually this well is within it’s limits. You might have a traffic light that says yes it’s green, no it’s trending up towards being not OK, or it’s red. Therefore the engineers only need to work on the wells that are trending in the wrong direction. And you can do that with almost any sort of data.
OGE: Are there specific areas that are very common in that regard?
Edwards: There are two or three common types of things that you do. One is production optimization. By having the live data and information you can optimize the volume from one end of the process to the other. We would see operations do an optimization loop once a day. Now, they’re not doing it once a day; they’re doing it as needed. And if its needed is four times a day, they do it four times a day. If it’s once a week, they do it once a week. By having the ability to do it more often, and indeed with higher quality because you’ve got the live data, you get production.
The second area would be around the condition of your facility. Instead of having a pump fail or a piece equipment fail and not be able to see it, now you can see a trend on that piece of equipment that says this pump is not looking very well. It looks like it’s going to fail but we’ve seen it early enough that we can do it in a planned way. We think the pump can get to a point where we can shut it down in a planned way rather than not finding out about it until it actually trips. So in essence you’re buying time, and what that does is it brings you availability in up time. So you’re able to keep your plant up and running, your wells up and running, for an availability increase by having the data and information and recognizing when issues are coming your way.
OGE: Let’s talk about the people and process sides of the business. You’ve got the people in the field and the people in the control room, and there usually are trust issues there. What are some successful strategies that you’ve seen for getting those two very different teams to trust each other?
Edwards: Well, there are a number of things that we’ve done pretty routinely. Typically offshore operations are land based. They’re left to their own devices. They inherently say, "We stand alone and we can fix anything kind of thing." So this is not about command and control and necessarily telling people what to do in the field. This about supporting the people on the field and saying we’re here to help you. If you pitch it in that way, it’s much easier and much more acceptable.
We run a series of workshops to say, "How is this going to change?" We transition the way that process is run collaboratively and remotely. There are some very simple things, like if the opportunity to rotate field people through the office, so that we say, "OK Joe, instead of going straight out to the field, if there’s cover for you, what we’d like you to do is come in the office this week and sit with the support engineers."
OGE: It’s about giving these teams a common vision.
Edwards: Having a common vision and a common organizational structure that says, "Here’s what we are trying to get to and where’s how are we trying to do that." Typically in land-based operations, operators might see a well or a piece of equipment every day, or certainly every week. And you get to the point where you use the data and information remotely to say, "We’re OK about only going there once a month or when we need to."
We’ve got the data and information; we can do the analysis. We can tell what’s in good condition, what’s not in good condition, we ultimately can prioritize our resources in a certain way. But that means that we’ve got to get the operators comfortable to a point of saying, "I’m happy and comfortable not to go there."
We were with one client, and we asked, "Why do you go to this well pump?" And they said, "We’ve got to check the corrosion indicator pump is running." And we said, "Well, can’t you see it’s running on the real-time data and information system?" They said, "Yeah, but sometimes the data is wrong. We don’t trust it." But they still want to go and see it.
OGE: So you’re right back to spending the money that you’re trying to save.
Edwards: Exactly. You’ve not only have you spent the money on the automation and you’ve not saved the money on the output. So the higher-ups say, "Well, why did we spend all of this extra money on the automation when we got no benefit out of it?" And that’s a very valid question.
OGE: Absolutely. So somehow you’ve got to convince these guys to believe the technology, to be OK with not visiting. I mean, these guys are mechanical and engineers and a lot of them just love doing it. You’ve got somehow convince them that that’s OK, through either that common vision or through just them understanding how the system is supposed to work.
Edwards: That’s right. And of course, many times there are some real issues. You know, there are some real issues. And then you’ve got to try and find a technical solution to those real issues, if you see what I mean. So, you know, you’ve got to take them on that journey and you’ve got to take them with you on this.
OGE: With the pace of this new process, do you see hesitancy on the part of the teams to jump up into this new speed? Do you see any hesitancy on the part of people charged with doing this?
Edwards: Yeah, that’s a really interesting question. We see this is very organizationally and culturally dependent. What we mean by that is that there are certain regional and national cultures that are resistant to this, and there are other cultures that are more accepting of this. Generally the older the operation, the longer it’s been online, and especially if it’s been told it’s the best in the world continuously the last 20 years, they’re the ones that tend to be the most resistant.
One of those surprising things that we find out with this is that if you put smart people in a room with the data and information and the ability to collaborate and the ability to understand what’s going on in the operation, things change quite rapidly. And as a result, processes evolve and change quite rapidly. So the cultures that do this well, are the ones who don’t see this as a one-off shot; they see this as a continuous improvement process. They’re always challenging themselves and what they’re doing, so their processes and their way of doing things is not particularly static. And it might be static for a year or 18 months or two years, but they’re always looking for the next improvement.
It starts out as a project because you’ve got to do something, you got to build something, you got to build a room and that data and information and do some IT, and do some change. But once it’s fully embedded, it continuously improves, or it should continuously improve and be self-sustaining. Where it doesn’t go well is it’s seen as a one-off shot and a project. And then of course, you know, the leader did it gets a pat on the back, gets moved on to bigger and better things, some of the team get moved on to bigger and better things, and then three or four years later you got a bunch of people sat in a room thinking "What am I doing here?"
The one thing I would say is it is important is that you do it, whatever way you do it, you do it that is aligned with your culture to an extent. You may want to change your culture, but you do it aligned to your culture. It’s very difficult to impose something on Lean to a culture that is opposite to that.
Original content can be found at Control Engineering.