How to Listen
Listen on Google Podcasts, Apple Podcasts, or via the SoundCloud media player above.
For many drivers, traffic jams have become part of the daily routine. Accidents, construction, debris and flooding can cause roadway slow-downs and traffic snarls. SwRI’s Integrated Corridor Management (ICM) system provides a solution, offering a fast assessment of highways and roads to find detours, reduce traffic trouble spots and get drivers moving. An operator works with the system to determine the best way to get around traffic messes. Drivers get more green lights and save time. As our guest explains, ICM is just the beginning of more connected, smarter cities.
Listen now as SwRI Manager Clay Weston discusses Integrated Corridor Management and how the system is mapping a new route for the future of driving.
Below is a transcript of the episode, modified for clarity.
Lisa Peña (LP): What happens when you encounter a traffic accident on your daily commute? No doubt you hit congestion and delays. An SwRI team has a solution to reduce traffic tie ups, turning more lights green and getting you on your way. We'll talk about it next on this episode of Technology Today.
We live with technology, science, engineering, and the results of innovative research every day. Now let's understand it better. You're listening to the Technology Today podcast presented by Southwest Research Institute.
Hello, and welcome to Technology Today. I'm Lisa Peña. Integrated corridor management, or ICM, is a collaborative system that collects data from roads and freeways. The system can react to issues, sort through options, and suggest ways to optimize the flow of traffic, especially after an accident or other traffic disruption. It's smart mobility that leads to less congestion and more green lights.
Our guest today is SwRI project manager Clay Weston, who has rolled out ICM in Florida and is making plans to grow the program. He's here today to tell us how this innovative system works, and what it means for drivers. Thank you for joining us, Clay.
Clay Weston (CW): Thank you, Lisa. Happy to be here.
LP: So I wanted to start with an overview for our listeners. What is integrated corridor management, or ICM?
CW: Sure, so integrated corridor management is a derivation of incident management that we already do. So we only work with Departments of Transportation to run their software that allows them to respond to an incident on the highway. So if an accident happens, we currently will tell a user or a traveler there's an accident ahead, two left lanes closed in two miles, or something like that.
Then they make their own decision. Maybe they use Google Maps, maybe they stay in the congestion because they think it's going to clear soon. They make their own decision about what they're going to do. Integrated corridor management takes that and moves it to the next level of actively telling people a different direction that they may take in order to minimize their travel time.
So we're actually going to change traffic signal timing patterns along predefined diversion routes, in order to give them more green lights in the direction they're going, get them off of the highway before they encounter congestion, and onto the highway after wherever the accident or incident occurred. And that way, you get less people entering congestion, more people going along the arterials, and getting those green lights to flush them down the arterials and back onto the highway.
LP: I mean, that sounds amazing. That's wonderful news for drivers. And it sounds a lot safer, because you won't have drivers sitting there trying to make decisions when they rushing to work. So I'm sure there are plenty of advantages and benefits to ICM. What are some of the top benefits you see with the system?
CW: So the top benefits is mainly just reduced congestion, and reduced travel time for people to get where they're going. There are lots of derivation of benefits off of that. You get less pollution, you get faster emergency response time, you get people with more green lights along the arterials, and then people get more confidence in the system so that they can use it more and more. And then overall as the system grows and we add more and more control, you get an even better control over the entire environment, thus providing the best mechanism of routing traffic around the city that you can.
LP: OK, so you are driving along, you encounter an accident, and this system starts giving you detours. Starts telling you the best route to get you away from the accident, and to avoid congestion. So how does it do that? How does it know what's going on around you? How does it work?
CW: Sure, so the first thing that happens is that we get notified that an incident happened, or the system gets notified. So an operator that is sitting in a traffic management center will be notified that an incident occurred. They will typically verify that with a traffic camera, or they will get phoned from the police or whatever, and they will enter an event into the system. Once that event happens, the system starts deciding what it should do about it.
So it goes to our system, which is called the ICM, and it then starts looking at what reactions can it take. So the first thing it does is looks at what is the congestion profile along the highway. So at the point of where the accident occurred, it says, this area is congested. It then starts looking up the highway, and it compares the current traffic speeds to the historical norms of that traffic in the area. And if it's within a certain percentage, then it will consider that area congested.
And it will keep going back until it doesn't find that any more, at which point it has defined the area that is currently congested. Because what we want to do, is we want to get people off before the congestion even occurs. At that point, it starts looking at the diversion routes. What are my options for getting people around this area of congestion? Where are the exits before the congestion, and where the entrance immediately after the congestion?
And it starts gathering those from the system, and then determining if those are viable. So one of the things we want to look at, is if the arterial that we are suggesting that people go down, instead of the highway, is already congested. Maybe there's already an event there, or maybe it's already over capacity because it's rush hour, or whatever the reason, there's already traffic. So maybe we don't want to send them down that route.
Or maybe we don't have control of those traffic signals right now. Maybe there's things that we know. That that traffic signal is in blinking red, or some other reason that if we sent traffic down that diversion route, that it would actually snarl traffic up worse than it was before. So we look at all these things and make sure that we can control the traffic signals, make sure we can control the message sign so that we can get the information out to people, in order to try to determine what the best route is.
We then send it to a simulation engine. The simulation engine takes in that diversion route, and simulates the traffic that will happen within the next 15 and 30 minutes to determine how the system would react overall if you were to implement that diversion route. And it scores that diversion route. And it says this would be better or worse than if you were to do nothing. And if it's better, then the operator can then decide to implement that diversion route.
At which point, it actually goes out into the field. We change the traffic signal timing patterns, we give people the relevant information so that they can make their decisions, and we start the process of actually implementing it and getting cars to make those diversion routes.
LP: So many steps involved. It's doing a lot. There are a lot of moving parts here. How is it gathering data from the roadways?
CW: So it's gathering data from all types of different detectors. We have traffic signals that we're getting data from. We have speed detectors along the highway that tell us how fast traffic is going, ramp meters that give information about how often people are getting onto the highway and stuff like that. We have DMS signs that are giving the information out to people. So we're collecting all this information all the time.
But typically, the whole event starts because there is an operator that has determined and verified that there is some kind of roadway congestion occurring, whether it be from an accident, or water on the roadway, or debris, or whatever it is. There is something that we know that will probably close some lanes, and cause traffic to back up. In which case, we might want to implement one of these diversion routes.
LP: What's great about the system is it does rely on human input, which is so important when you're talking about traffic because things can change so quickly. So tell us about this human interaction. Why was the system designed to rely on the human input, as well as the mechanical?
CW: Well sure. One of the things that we encountered very early on is that the Department of Transportation is responsible for the highways, whereas the local agencies are responsible for the traffic signals in their jurisdictions. And so in order to route traffic down these arterials, we need to get concurrence from these local agencies that it is OK to do so. So to do that, whenever one of these comes in and we determine what diversion route might be best, we have to poll the agencies and make sure that they approve of doing the diversion route. That they are OK with you changing their signals, giving the messages to commuters to go different ways, and stuff like that.
And it goes out to them so that they have approval and they have knowledge of what's going on within the system. It's very important that everyone knows what's going on. If there's going to be additional traffic routed down their roadways, they want to be able to know about that so that they can coordinate other responses. And so that emergency responders, and stuff like that, can know what's going on. And so it's all about information sharing and gathering between different - the Department of Transportation, and the local agency users, and then the operators that are actually actively managing the event at the time.
LP: All right, so we've talked a lot about accidents. How the system deals with an accident occurring. But can the system handle other traffic disturbances?
CW: Sure, absolutely. So if an operator decides that there's something going on on the road that will cause a disturbance or that needs a response, whether it's a tow truck, or whatever, that needs to go out and clear debris. Whatever it is, if that's going to cause a traffic disturbance, then they will probably enter that into the system.
And once they enter that into the system, that's when the ICM starts chugging on the data and deciding if it should make a diversion route or not around it. So it reacts to all different types of inputs, it just depends on whatever the operator decides is something that will actually impact traffic or not.
LP: What type of roadways work best with the system? Does it work better on a highway, or does it matter if it's a country road? Is there any difference in how it works, how it functions?
CW: Yes, absolutely. It is designed for use on major roadways in city areas, because that is where you find your most congestion. And that is where you get your biggest snarls of traffic whenever an accident is occurring. So it wouldn't really be designed for a country road, or a rural road, or anything like that, because of that.
And another reason is that it really only works in areas where there are viable diversion routes that have the capacity and the lanes to allow traffic to go along them, and not stay on the highway. So another reason that it works majorly, and are mainly in city areas that have good arterials and diversion routes that can go around the area of congestion.
LP: So where is the system currently in use? I know we said at the beginning you rolled it out in Florida.
CW: Sure, this just started being in use in the greater Orlando area in Florida. It's the District 5 of the Florida Department of Transportation, is the product that we made this for. And so it's the I-4 corridor that goes, it's an East-West highway that goes primarily North-South, because it's Florida. And it's the Orlando-DeLand area. It involves Volusia County, and Seminole County, and Orange County. And that whole area in and around Orlando, focusing on the I-4 corridor.
LP: And how long has it been in use there?
CW: Only a couple of weeks. We just rolled it out a couple of weeks ago. It's in its first time where the operators are getting used to it. They're starting to use it, and starting to become accustomed to it. And they will start using the diversion routing soon.
LP: All right, well how's it going so far?
CW: So far so good. Like I said, we're still in the early stages, but early results look good. The amount of - there's a lot of factors that go into tuning the system. Making sure that the area that you consider congested is accurate, so that the speeds that are under historical norm, making sure that that is an accurate representation of what qualifies as congestion is very important. Because you don't want to send too many of these.
You only really want to do it whenever it's a major enough event to want to get people off of the highway and onto the arterials to get around it. Because if you do it too often, then it's not going to be as beneficial. You're going to snarl traffic up along the arterials worse than you would on the highway if you were to just let it go.
LP: So you have a contract with Florida, is that correct?
LP: OK, so, while you are actively putting this out there, it's also become a bit of a learning ground, I imagine. Where you'll take what works there and use it in other areas. Is that is that the plan right now?
CW: Yeah, absolutely. We're currently under contract with the Tennessee Department of Transportation to do an ICM for them, where we are using some of the lessons that we've learned from the Florida project and bringing it to that system, in order to enhance it and make it even better. And over time, as we do more of these projects and more of these efforts, we will continue to learn to see what works and to make it better over time.
So there are a couple of lessons learned that are important from that. One of them that came early on in Florida is that well, there were two ICM projects before us, that we were not involved in. One in San Diego, and one in Dallas. And one of the things that happened with those projects is that it took a really long time for the agency operators, at times, to respond to an event. And so what happens if you don't respond to these events quickly, then congestion just keeps getting worse and worse. And then any diversion route that you implement isn't as effective.
The quicker you can do it, the better. And so one of the things that we implemented for Florida is the idea of approval profiles for both agencies and devices. So what you can do with our system, is that I spoke to the ability of an agency to go in and approve the change of a traffic signal timing plan. Well what they can do is they can set up an approval profile ahead of time that says between these two times, they will automatically approve a traffic signal change after, say, 10 minutes.
And that gives them the ability within those 10 minutes, whenever this occurs, to go in and look at the system and determine if there's something else they want to do, and then make a decision. But after that 10 minutes, it will automatically approve for them. And this ensures that the system is actionable and operable in a very quick time frame. So that you're not waiting on users responses all the time in order to implement these diversion routes.
So we're taking that into consideration, whenever we do things for the Tennessee Department of Transportation, TDoT, and we will incorporate some of those things into their system. Another thing that we did was that the Florida system was originally going to be based off just of the severity of the incident. So there's a severity of one, two, and three - minor, major, and critical - and depending on that, we would select certain diversion routes that go around those accidents.
We made a pivot in the middle of the project to go based off congestion. Because we figured it was more important to know where exactly the area of congestion is in order to get them off of the road, instead of just the severity of the incident. It's a more granular level of information that allows us to make more intelligent decisions. And we will continue to take those things into account as we build more of these systems and make the technology better and better.
LP: Wow sounds just, kind of a really great-- so many great advantages for drivers, having the system in use. We're here in Texas. I'm wondering, when will we see it here? Or when do you expect to see more widespread use?
CW: Well that's a good question. It's caught on across the country. There are multiple grants, federal grants, that have gone out for ICM systems. And so there was an initial one done in Dallas. I'm not sure if that is still currently in use or not, but it could be done. I would love to see it here in San Antonio, our home base for SwRI. And we have worked with the city to do different things, so I'd love to see it here.
And I think as it becomes more commonplace, and especially as we get more and more data, we will be able to do this more. One of the things that caused us delays in the system is just the gathering of all that data. You have to really know what is going on in the roadway. You have to have everything properly instrumented, and the data that comes from those instruments pulled into the system in order to use it.
And more and more cities are getting there, and becoming smarter and smarter, and adding more detections. And so as that happens, we will have a better idea of what is going on, and we will then be able to better control the different devices that we need to control for these kinds of systems.
LP: How long was SwRI's ICM system in development?
CW: It was in development for about three years. And part of that, it was originally going to be a shorter time frame, but part of that was due to the amount of data that we needed. There were traffic signals that had to come online, that we needed to integrate with, that took a little longer to get there. So it took about three years for us in development. And then we are just, like I said, just going live these past couple of weeks.
LP: So how does it developed? You mentioned the sensors were in place, cameras have been in place. But when did you decide, how did you decide, we can do more with these things? We have the capability here to make this even better. How was the system-- how was it developed?
CW: Well, so it's probably important to say that Southwest Research has already developed what we call an Advanced Traffic Management System that is in use in Departments of Transportation across the nation. I think we're currently in 13 or 14 states, where they use it for their main software for managing accidents and managing roadways for the Department of Transportation. And so we already have the experience of knowing what is going on with the roadways, of knowing how these systems work, what works best, and what can be done better.
And so the Florida Department of Transportation, I think in 2017, they originally did the request for proposal for this, which we responded to and won. And one of the reasons for that is we had already written their SunGuide software, which is what their operators use in the TMCs in order to manage their roadways. And so this was just a continuation upon that. It's a separate system, but it integrates very closely with the SunGuide software, in order to share information back and forth and make these response plans go into it.
So we continue to enhance all of our offerings as we go. Obviously, we are contract for hire, so when a state. DOT wants to do more, they can come talk to us and talk about what is important to them. And we often bring ideas to them too, about what can be done better, and how we can implement new changes, and things like that.
LP: So what has been your biggest breakthrough, or most exciting moment, in this development process?
CW: I think probably seeing everything work in the end. Our system is so dependent upon all of the different data pieces that come in, like I said before about, from traffic detectors, and signals, and all of these things. And to see the complete flow through of getting the data in from all of that, towards grabbing response plans from the system, and talking to the simulation engine that we talk to in order to score them; all the way through getting operators to select response plans, sending those response plans, and seeing them actually enacted in the field.
That was probably the biggest part of it. And like I said before, there were multiple points of where we found how we could do it better. Like when we switched to going off congestion instead of severity, and things like that. So that's what I would say.
LP: What is your - we've talked about this a little bit already - but what is your long-term vision for the system?
CW: Sure. Long-term is smarter cities overall. I think eventually, the idea is that cities will be largely controlled via software. All the traffic signals, all the different detections, and ramp meters. And connected vehicle information. All of that will be pulled into a central repository, and can then be acted upon. So we're getting more and more into the area of connected vehicle. As part of this project, actually, we are sending connected vehicle information out to cars that do have that capability, in order to give them on-screen and in-dash notification of what is going on.
And then especially as we get to automated vehicle, this information will be especially important to automated vehicles, so they can immediately take response and action whenever these things are implemented. And then just greater traffic control overall. In Tennessee we're getting into - we're looking at using variable speed limits along areas to better control the flow of congestion in and around incidents.
We're also getting into dynamic lane control. So dynamically being able to turn on and off lanes where people are traveling, in order to affect the way that they move around, and go around accidents, and things like that. So overall, we will just be getting smarter and smarter, and able to have more control over the different things within the city that we can control, in order to route people around the best that we can and get them where they need to go quicker and safer.
LP: I mean, can the system work during normal rush hour congestion? I know you mentioned it's very focused on an incident occurring. But, gosh, can you see the day where maybe you can eliminate the rush hour? Just asking, wondering if it's possible. We've come to expect congestion, but I'm just like, imagine if we could just easily get through where we need to go because of ICM. I mean, could that happen?
CW: You'll never see me use the word "eliminate." We will never eliminate congestion, unless every road has 12 lanes, and everyone can get where they're going however they need to. What we can, is minimize it as much as possible, given the infrastructure we have. So what we found with cities, is that there's less and less ability to expand outward, and add more lanes. But what we can do is add more infrastructure and technology, in order to route people around more intelligently, and get them where they need to go quicker and safer, like I said.
So I don't know that we will eliminate it, but we will certainly make it better over time. And it will be a increase over time, so that congestion will get better and better. But then there's other factors about how many people are moving into the city, and other things that affect congestion. So it's an ongoing battle, about how we can make things better, while dealing with the amount of cars and traffic that we have.
LP: All right well any improvement is a good improvement, so we'll take it. That's great. So can the ICM system be expanded for other purposes? are you looking at other ways to use it?
CW: I think, right now, we're primarily focused on getting it out there. What we want to do is we want to make it ubiquitous, right. We want to make it where everyone can use it. Where it's everywhere that we have control over these devices, and we have the information necessary. That we get that control, so that we can actually perform these actions. And then, like I said before, we can get better and better.
In Tennessee, as I mentioned, we're doing variable speed limits, we're going to be doing active lane control, those kinds of things. We're doing intelligent ramp metering, right. How soon we'll get cars onto the roadway. So there will be other things that we get into, and other means of control that we gain over time, and we want to use those in as many places as possible. Because that's the most important thing, is as we can do better, we want to get that out to as many people as possible, so that they can make use of it.
LP: Looking forward to that. So I wanted to ask about you personally, Clay. What is your professional background? How did you get onboard for this ICM project? What kind of preparation did you have to make these decisions around the integrated corridor management?
CW: Well I have a computer science degree from Trinity University in San Antonio, and I initially went to work, for a couple of years, at a different company before I came to Southwest Research in 2007. Since coming to Southwest Research, I have worked in intelligent transportation systems. I worked for the Central Florida Expressway authority contract, which is another tollway in Orlando. So I was already familiar with the area, and I was already familiar with how we write software in order to better handle traffic, and stuff like that.
I also worked on the advance traffic management system that we use in Texas, called Lonestar. And so that gave me the knowledge of how these systems actually respond to incidents. How we actually currently route traffic and things like that. And from those years of project manager experience, and software experience of how to build these systems, it led naturally into helping create these systems for ICM, and making it better over time.
LP: I like to ask that question for college kids looking for a career path, and this is really interesting. This is probably not something you hear about often, but you guys are back there making these big decisions around traffic management. And that affects all of us every day, so it's just wonderful that the team is there and doing this type of work. And I like to inspire other people looking for a career path that this is out there. Something to think about.
CW: I'll tell you, coming out of college, I certainly never thought my career was going to be in transportation. I came out of college with a computer science degree, thinking I was going to make video games. Probably as a lot of young college students do, right? But getting into traffic, you realize that there's a lot of really interesting problems. And that's what computer science is, it's using software to solve problems.
And so traffic, there's so many interesting ways that you can use this information. And you can write machine learning algorithms and big data algorithms to gather all this data and make intelligent decisions off of it. There's some really interesting problems and concepts that you get into in transportation, that I never thought I would really have access to.
LP: Yeah, really intriguing that all this behind the scenes work going on to make our driving situation a little better. So I did want to end with the big take away today. What would you like our listeners to remember about the integrated corridor management technology?
CW: I think just that overall, State Departments of Transportation, they're looking to make your commute better. We want to make your commute easier. We want to make it quicker, and we want to make it safer, so that you can make intelligent decisions about where to go. And you can have the information that's relevant to you in the time that you need it.
Because the idea is that the quicker we can get people where they need to go, the less accidents we have, the less congestion we have, the less pollutants from just driving we have. So it's an overall benefit to make the cities safer and smarter, and get people where they need to go quicker.
LP: Simply amazing. As all of us, we're on the roads every day. This is just a really great system, and your team is just-- the work you're doing is astounding. So we have learned to live with slowdowns on our commute, but ICM means, in the near future, those slowdowns could be reduced. We may not have to live with them every day. So that is life changing, and all I can think about is all the time we would get back if this was in use everywhere.
So we applaud you and your team, and thank you for sharing your work with us today, Clay.
CW: Thank you so much. It was a pleasure talking to you.
And that wraps up this episode of Technology Today. You can hear all of our episodes and see photos and complete transcripts at podcast.swri.org. Remember to share our podcast and subscribe on your favorite podcast platform.
Want to see what else we're up to? Connect with Southwest Research Institute on Facebook, Instagram, Twitter, LinkedIn, and YouTube. And now is a great time to become an SwRI problem solver. Visit our career page at swri.jobs.
Ian McKinney and Bryan Ortiz are the podcast audio engineers and editors. I am producer and host, Lisa Peña.
Thanks for listening.
The innovative technologies and novel strategies leveraged through integrated corridor management (ICM) help address congestion on major metropolitan roadways. A leader in intelligent transportation systems (ITS), Southwest Research Institute has over 20 years of experience developing, deploying, and maintaining small- and large-scale ITS systems. We deploy unique ICM solutions as systems integrators, software developers, and R&D problem solvers.