The Roundtable 2024: New Voices for Change

“So, how much of the barrier to a quantum leap in industry performance is due to lack of tech? How much of it is due to resistance to change? Can this industry really adopt transformative technology fast enough if the process flow does not change?” – Jemmie Wang

DEVELOPMENT, DEPLOYMENT: QUESTIONS EXPLORED
For the first time, the Common Ground Alliance partnered with Planet Underground TV to bring The Roundtable to its Fall Damage Prevention Summit in Las Vegas, Nevada. We welcomed many new voices to these important conversations about the direction of damage prevention going forward.

Topics covered in this table included, what new technologies/methods can be used to better manage our crowded underground utility infrastructure rights-of-way? What groundbreaking technology can make finding underground utilities easier? Can AI help the underground utility industry in an impactful way? Is it already happening, and could it present problems as in some other industries? What are the barriers to the industry leveraging technology more?

This transcript, taken from the final panel of the event, is one of many conversations (edited for print) from The Roundtable. Visit our YouTube channel to watch all of this year’s Roundtable discussions in full.

WHAT FUTURE TECHNOLOGIES AND METHODS ARE YOU WORKING ON AND ARE ON THE HORIZON FOR DAMAGE PREVENTION?

Shane Hart: We talked a lot about APIs this weekend, application programming interfaces, in which computers talk to each other just like humans do. They use a language and certain protocols, and as we’ve seen with Texas 811 where they bring more data into the call center problem and have more data to cross-reference, they get better solutions, better decisions.

We already use the APIs for Esri to pass GIS data to different platforms in real time. The call centers need to follow a similar process, where that communication is almost instant from the front end and the back end, so that we’re feeding the most data that we can have to make the best decisions. Right now, it’s a little bit of a bottleneck.

Shane Hart (on left)

Jason Manning: In my mind, the data collection should already be happening, but I really think it’s going to be key to the future. We’ve got all these emerging technologies, and I keep hearing the same thing—we don’t have great data. For a lot of utility companies and facility owners to not have their data is like an e-commerce company not having a database of customers. So, that’s really got to happen to enable a lot of the other tech that’s coming down the road, so that we stop hearing, well, we don’t trust our data. There’s so much going in the ground now. Collecting the GIS data is huge.

Steve Mumm: I would add to that—using different collection devices in combination with one another to create even more rich data sets. Tools that we haven’t classically used in the utility locating world can provide access to things that are deeper and access or coordination with things that are above the ground.

Jason Adams: The one thing that is really ringing a bell for us is the ability to use the data and get the value of the insights out of that data. And what we’re starting to see is an expansion of the conversation, right?

When I first started, we were very much just looking at a ticket. Today, we’re looking at the positive response and all the different activities. So, what we’re doing is introducing a lot more data into the conversation. I think that gives us a lot of alternate ways to get value out of the relationship.

I’m super impressed with all the new organizations that are coming into our industry, because we’re now seeing people invest more money. And I think that’s because we see the opportunity and the value of what’s here. I also think some of the new tech that we’re seeing is really going to change our industry in the next five years.

Mark Whelan: I think we’re in a golden age, really, of technology. I mean, cloud storage 20 years ago wasn’t as much of a thing. Now, it’s so cheap and so readily available, it puts the power of information in the palm of anyone’s hand. So, I think collecting all that data becomes a lot easier when you’ve got a place to store it, and we’ve got the ability to transmit that data anywhere seamlessly and quickly where it needs to be.

Jeff Hancock: And importantly for me is the validity of this data that’s coming in, how it’s collected, where it’s going, the accuracy of it, and how you verify it. Some of these smaller municipalities that don’t have accurate maps and records, what does that data look like?
Shane Hart: Even the small municipalities, they might not have resources, but it doesn’t mean that they don’t have these problems too. They need to find creative ways to get that data any way they can.

Jeff Hancock

For a lot of people, it’s got to be high quality data, but that’s what the AI does later on. You just feed it everything. The data already exists out in the world. The software and the processes are just wrapping that data so we can track it and learn things from it. The AI will tell us what data correlates.

When you do stats, you have longitudinal studies. Doing something today, we’re not going to get the most insight. We might not even know what happens in five years, but if we keep collecting this data every time? At some point, we’ll get to where we’re using GPS every time, because I might want to know what the path was the operator took. I want to know the swing quality from the Vivax every time. At some point, you have a time lapse picture, and you could keep states of your database with valuable information.

Jason Adams: The one thing I always struggle with is, when we come to our conferences and do things like this, we’re always talking about fixing the problem that’s kind of five feet in front of us. But at least in my experience, what I’m starting to see is, I think in the next three to five years we’ll find repositories of data that are more accurate than the one-call centers will be able to produce, because we‘re already showing ways in AI how to create those layers of very specific GIS without needing GIS. And so now what we’re saying is, the one-call centers that have been so valuable as a central repository for so long still have value, but we also have to be aware that sharing that data, especially when it’s being created in outside sources, has a lot of value.

I think to hit on the point, making sure that it’s valid data, I think is really the key to the conversation. If we start building these great ways to anticipate making better decisions, but we’re not sure the fundamental soundness of the data, then that’s the conversation we really need to get into.

Jason Manning: There’s probably a big shift that needs to happen in the way that data has been protected. And there have been, at least in my experience, some gatekeepers to that who say, I don’t know, this is sacred stuff, can’t update it, can’t change it, can’t do anything with it. It feels like that whole mentality is just going to get passed by.

I use this example all the time when you’ve got anything that’s crowdsourced. It doesn’t matter if Google Maps says there’s an Arby’s down the street, if 50 people follow that there and they go, hey, this is closed. Well, at some point those 50 people are probably right. It doesn’t matter.

Jason Manning (center)

There’s no feasible way for them to send someone around checking every single thing. And that’s kind of the mentality that a lot of the industry has, as opposed to going, hey, any given spot in any given city has probably been located over and over and over again in the past 10 to 20 years. At some point, those mistakes, if we collect all that information, are just going to sort of update themselves. We’re not going to need these gatekept GIS repositories. They’re going to get left in the dust.

Jason Adams: I’m sure you experience it. We experience it now running our one-call centers. They’re constantly being attacked to pull data out. We never saw this three years ago. We never saw anyone trying to log into these systems to pull ticket data. And now we’re seeing it in every one of our scenarios. So, we’re spending a lot of time trying to build walls. But what we’re really hoping is that we can start building bridges and really start sharing this data and say, hey, there’s real value here, instead of trying to build walls, where we can’t patch the holes fast enough.

Shane Hart: A lot of the time, when I bring on customers or I’m in meetings, there’s a big component of software education that you’re providing as well as the service. When we talk about the APIs, and when I talk about Esri, we have GIS. I have people that say they’re GIS experts, and they’re nervous around the API. And they think giving you a stale copy of the data that you can load on your laptop that they’ll forget about—and that laptop could be anywhere—they think, well, that’s a lot safer. And it’s not.

What happens in these API protocols now, if someone’s accessing your Esri, and we do this with some of our big utility customers and our vendors, we’re pulling that data in real time to the locator so that they can view it. And we’re using that data to do things like pull a high profile before it’s on the locator. We’re telling the locator beforehand, hey, this is high profile. Can you tell us? Yes, no.

So, we’re setting up an environment where we can crowdsource that. But the Esri on the token generation is way more secure than if you’re passing stuff around manually, because then you have timestamps, you have logins, you have passwords, and you have token generation.

Mark Whelan: Back to the validity point, in the past, we’ve relied on interpretation of data. And I think we’re moving towards a model where the collection of the data can actually be determined. Basically, the person that’s collecting that data won’t have to be interpreting something.

So we’ll know that the data is valid and is accurate because the machine, the locator itself, will be able to collect enough information to determine the position of the actual utility. And when that gets fed into the system, you’re going to have much more accurate maps rather than what happens in a lot of cases. The locator goes out there, collects points based on his interpretation, and that interpretation may be wrong, it may be mismarked, and then you have a wrong map. And then you get into a whole myriad of issues around that.

Jeff Whelan (on right)

Steve Mumm: I think there’s power in combining data sets as well. We just came out of the best practices committee where I talked about combining a private utility locating set with a Texas 811 data set. Just to give more context to that excavator who’s on the ground, those utility sets do come together. And they’re not mutually exclusive. Providing that context materially impacted the dig up rate for those contractors. I think the more we can experiment with how these combinations of data can affect outcomes, the faster we’ll be able to move the industry.

“Someone has to be a first mover. Someone has to take some risk.” – Steve Mumm

Jemmie Wang: So on that point, we come to these conferences and one of the most common themes is, we need to share our data more. Steve, what you talked about seems to me to be a perfect example of sharing data. You guys went out and you know there are private facilities there that overlap with the dig area, and you’d like to share it with the public. It seems like a no-brainer. It’s talking with some stakeholders. Most are supportive, but there’s some trepidation. Isn’t that a great example of sharing actionable data in this industry?

Steve Mumm: I think it’s on us to make some of those bold moves. Marathon did it with their pipeline data a few years ago. We’re trying to do it more with our data sets. Someone has to be a first mover. Someone has to take some risk for no other reason than to experiment and pilot with what that data can produce.

Shane Hart: It’s very similar to the abandoned pipes too. Someone said it took 10 weeks to find out who wanted to take responsibility of that pipe. And it’s another data set that affects the excavators when they’re out there trying to work safely. I think the call center is coming to where we’re servicing the excavators to make sure.

That’s why adding those things, whether it’s private or abandoned data, is helping to create a fuller picture for the excavators. Jemmie, you talk about the excavators, there’s something they can do. But maybe we can incentivize sometimes to help crowdsource. Because at some point everyone has some data. You do the job today, and that’s great. But if you collect a little bit of data every time, there’s a play down the road for you to add value on what you’ve been collecting over time. And everyone will be able to share their data in a safe method.

Steve Mumm (center)

Jason Manning: Even for the abandoned stuff, we’ve talked about this before, where, as you said, if everyone is out there doing it, and people find abandoned facilities, if they collect that data, if that’s shared in a way, the technology is here now to share that data in a much more controlled way. It’s there. It exists. It’s around.

I think there are just a lot of people hitting the brakes on it because they’re concerned. But there are so many things that can be done if we just, as you said, get some people to start doing it, to start sharing abandoned lines, to start sharing the data that’s been crowdsourced. People will realize it’s not going to be the end of us, and that it actually can work and lets us move forward.

Jemmie Wang: So, how much of the barrier to a quantum leap in industry performance is due to lack of tech? How much of it is due to resistance to change?

Jason Manning: I’m going to say resistance to change. At some point, hopefully, we’ll catch up and then the bottleneck will be tech. But I think right now, there’s so much tech that exists and the industry is not adopting it.

This is my industry, and I love it. But we’re talking about an industry that is still struggling to get rid of fax machines. Because there’s a great resistance to change. Not all that long ago, call centers were resisting doing tickets online. They thought, well, it’s not reliable. We can’t do that. That’s dangerous. Why would you let people do that? And now, virtually every one-call in the country has the vast majority of the tickets coming in that are self-serve.

Jason Adams: Tech’s not the limiting factor. Resistance to change is, and finance is, right? I mean, I think we rarely have this conversation enough in our industry, that we have this great goal, 50% reduction in five years. We have really good strategies, but one of the most important strategies is the financial strategy.

How are you going to do it? What’s the incentivization? We want to be safe, but for vendors like myself and others, we want to build really good solutions that give good returns.
One of the things that we’ve really seen now is the value of the data in our industry. It’s starting to demonstrate value outside of our industry. And that’s why we’re seeing new people come into our industry. That’s great, because now they’re bringing new tech in. We have to educate our audience that we’ve had these relationships with for 20 plus years to say, this is beneficial, we have to invest in it. And I think we need more people to have that conversation.

We’re really trying to push the envelope as hard as we possibly can, tying AI to all of our products. But the reality of it is, until we get an audience that accepts and sees the value out of that, we can’t get a whole lot of change.

Mark Whelan: I think there’s a case for the technology companies to make sure that they can show the value of what they’re doing so it’s not perceived as something of diminishing returns where it’s like, well, it’s going to cost me a lot up front. How much more is it going to gain me in terms of damage prevention? I think it’s incumbent upon the technology manufacturers to basically present the case and develop use cases that actually deliver on what is the actual value that’s derived from this.

“ I don’t think there’s ever been a scenario where someone said they would like to make a decision with less information.” – Mark Whelan

Jason Adams: We should be working together as an industry. I mean, look, we have so many boots on the ground, and we collect great data, and the value of our industry data is incredible. We need to be working together to do two things, validate that it’s great data and know what we’re doing with that data.

If we do that, for us as an industry, it has a ton of value, but if not, we’ll let other companies come in from outside. We see this already, trying to recreate our data. That’s taking the value away from our industry. That worries me, because I want the value of our industry to reduce damages. That’s why I’m here. That’s why I’m working so hard. What I don’t want is somebody to come in and say, hey, look, I created this great data set which just duplicates what we’ve done here. And I’m going to use that value outside of the industry and remove the value from it.

Jason Adams

Steve Hart: You build software for locating. There are not enough people. There are limited resources. There’s no way to get all that work done. And you say, what other industry are you building software for at this level? And when you build the software, you build some very advanced features. We’re seeing excavators say hey, we need a good TMS platform where we can have GIS integrations, and they’re still doing the call center, but they also have access to the publicly available maps that we’re pulling in real time for them. So they can use that as a double check. When you do stats, you do correlations and you run some regression models, and you get confidence coefficients where you’re 90% confident at this level that this is a correlation, or I’m 95% or 99%.

In the technology committee, we’re doing more case studies. And it seems like we spent a lot of time of trying to show how this stuff can keep being better. We still have to carry on. Someone said the other day, I still have to put good paint on the ground. That’s true, but you need to start also looking at tomorrow.

Steve Mumm: How we define our industry, I think, is important. It’s not utility locating, let’s call it construction. If we’re part of the construction industry, we’re identifying information at a specific point in time in the life cycle of a facility when it is about to be exposed. But years before that, it was installed. If that information can be used for future planning for other projects, and then brought back to prevent damages the next time it’s exposed, the data can really be helpful for all stakeholders in the construction process well beyond just this moment in time that we’re so laser focused on right now.

Steve Hart: We see that with Jeff and On-The-Spot, some of that collaboration. We see this with CenterPoint having issues with the map. All of a sudden it’s an email to everyone. And everyone’s working at that problem together with the utility, the locate company and the software vendors, which is really nice.

Sometimes you don’t have to go out and track, you know, GPS and do everything to the pole. Sometimes we need to say, we’ll pose the question, is this a map discrepancy? Yes or no? And the next locate that comes in can cross-reference a dig area to see if there was any doubt on mapping in this area. Right?

So even if we don’t have limited resources, we need to always keep trying to get better with whatever we have available. Sometimes people think, I don’t have the budget, and I’m stopped, but there are some creative ways that you can really make some improvements.

Jason Adams: We just sat in a meeting about DIRT, and we looked at all this great information from 2023, and it’s 2024 right now, and we’re trying to figure out how to prevent damages. So a damages database in real time would be one of the most significant contributions of this industry I think. Because then we could start really start trading models on real time data. Right now we have to reserve all of our activity on building these AI models until a year later. It puts us behind, and there’s no reason this industry can’t have real-time reporting.

Jason Manning: The speed that these things move at, if you think about feed grants and broadband construction, it tears through a state like wildfire. It’s good and bad, but the heads up that you would get with real-time data would be extremely valuable in the prevention aspect, not just in retrospect, as we look back and we see what happened last year. You’d actually be able to get ahead of it and see what’s coming, where it’s coming, what’s going on.

“… the heads up that you would get with real-time data would be extremely valuable in the prevention aspect, not just in retrospect, as we look back and we see what happened.” – Jason Manning

Jason Adams: Imagine if there was a damage yesterday out on the line and you go, well, maybe I should adjust a little here, right? Because if you’re controlling the one call center, if you have utility software, if you have contract locating, we’re all talking the same.

The thing is that we build these walls to prevent us from communicating the most proper information that would reduce this activity. We have to get better at working together and sharing information, but we have to know how to do that in a way that it protects our clients, but it also leverages the advancements that we have because there’s a lot of knowledge at this table.

What I don’t want to happen is people that are not damage prevention experts telling us what the AI should say. I like the fact that I look at it and I go, I know I’ve got 30 years of this. I know that’s wrong. We got to figure out what’s wrong with the form. That’s why I hope we can still accomplish a lot of that while some of us are still in this industry. I’d hate to see a whole group of people come in without a level of expertise and tell us what the AI should say.

Mark Whelan: I don’t think there’s ever been a scenario where someone said they would like to make a decision with less information. So I think the more information we can gather and collate and leverage AI to basically churn through it, because obviously there’s more data being generated at any given time today than there ever has been in the past. And we need a method to mine through that information and pick out actionable insights. And the only way to do that is to leverage things like AI and that type of platform.

Jason Adams: You hit a perfect note. We have to know what questions to ask for the new data. We’re so fortunate now as we start to expand on positive response and everything else that we’re doing, and we’re introducing a lot of new fields of data into the conversation. My question is, what aren’t we asking? What fields of data would be valuable?

Only two tickets in the entire United States has a field for: I’m done with my ticket, or I’m not. Not a closed quote, but just, I’m done. You know how much it would be effective for our contract locating clients to know, hey, they’re done with excavating, and we could pass that information along, but only two call centers do this.

Jemmie Wang: Great discussion. So, there are two gentlemen on this panel, Jeff and Steve, who are actually in the field doing damage prevention. They’re where the rubber meets the road, right? So, let’s start with Jeff. How do you use new technology, you and your team, in the field? What has worked well for you and your clients? Talk about new technology or methods. What have you and your team been able to leverage most?

Jemmie Wang

Jeff Hancock: I know we’ve experimented with some of the RTK satellite imagery mapping. It’s fairly new to us, so I’m just kind of developing our own accurate map.
When we do the private side of utilities, we try to do that for our clients, develop those mappings with that type of technology, and integrate that into some of our public side. When you’re talking about trouble locates or locates that maybe you have to expose, we’re daylighting those to see where they’re at, and just mapping those for the future so that the data is accurate. And then you can go out, and next time it’s accurate, so the next locator knows, hey, there was an issue here. It’s mapped. It’s documented. It’s accurate.

Steve Mumm: I would second that on accuracy. RTK is just one tool, but the combination of new tools really does help close the gap between each tool.We talk about it like Swiss cheese, right? You’ve got to have the right tools to make sure that you don’t have any of those holes in the Swiss cheese exposed. Second, I would say completeness, doing larger scan areas so that you have more context about that whole area.

A full site, a full university, for instance, gives you a lot more information than just one area where they may be looking for a sewer line to install a chiller so that they don’t intersect with that. The last thing I would say, is that the more that we can think about that, not just GIS, but in context more broadly, whether that’s part of BIM and a BIM model. We’ve had some really cool projects where we’ve scanned a concrete slab and found all the post-tension cable and then looked at that in context with the building plan for the bathrooms.

You think about a 12-story building, and those plumbing lines are going to flow all the way through 12 stories. If you know where the post-tension cables are, now you can design around those interfaces. I think that’s just an example where we can go beyond just a utility locator and do something really great with a combination of data.

Jemmie Wang: Can this industry really adopt transformative technology fast enough if the process flow does not change?

Here’s a very quick example. Many years ago, we were looking at doing meter readings for a utility. They have routes every day. I say, okay, what if we figure out a better way, a more efficient way to read these meters? Can I do two days in one day? They said no, you cannot change this route. You cannot change a sequence. You cannot change the number of meters you read a day. You cannot change anything. So I said, basically, you don’t want me to improve.

To listen to the rest of this great conversation, plus all the rest of this year’s Roundtable discussions, visit our YouTube channel. Please comment and like the videos. We’re always happy to hear your views and gather more important topics for future events like this.