March 20, 2012 Active Traffic Management (ATM) and ... 20... · Web viewNational Transportation Operations Coalition Talking Operations Webinar Series Active Traffic Management (ATM)

National Transportation Operations CoalitionTalking Operations Webinar Series

Active Traffic Management (ATM) and WeatherMarch 20, 2012

Jocelyn BauerHello and welcome to the NTOC Talking Operations webinar on Active Traffic Management and Weather, hosted by the National Transportation Operations Coalition (NTOC). I will be giving a brief introduction to the web conferencing environment before turning the session over to our first speaker. Today's session will last approximately 90 minutes, with 70 minutes allocated to the presenters and the final 20 minutes for audience question and answer. Please be advised that our seminar is being recorded. During the presentation, if you think of a question, you can type it into the small text box in the chat area on the left side of you screen. Please make sure you send your question to everyone rather than just the presenters. The presenters will not be able to answer your questions during their presentations, but they will be able to address most of them during the question and answer session in the last 20 minutes of the webinar. A file containing the audio and visual portion of the webinar will be posted on the NTOC website shortly, and I will type that address into the chat box.

Attendees will be notified by e-mail of the availability of the presentations, recording, and closed captioning of this webinar online. We encourage you to direct others in your office that are not able to attend this webinar to access the recording online. The presentations used today, with the exception of Roemer’s, unfortunately, are available for download in the file download box on the left side of your screen. To download a file, click with your mouse on the name of the file that you would like to download and then click the button on the bottom of the download box that says “save to my computer.”

During today's webinar, you'll hear five presentations on the topic of active traffic management (ATM) and weather. First you'll hear from Dr. Roemer Alfelor from the Federal Highway Administration (FHWA) Office of Operations with an overview of Weather-Responsive Traffic Management. Then we will hear from Kevin Balke with the Texas Transportation Institute and Deepak Gopalakrishna with Battelle on Weather-Responsive Traffic Management using ATM Strategies. He will set the stage there. On Variable Speed Limits on I-80, we will hear from Vince Garcia from the Wyoming Department of Transportation, and then we’ll hear from Brian Kary with the Minnesota Department of Transportation on ATM in Minnesota during Weather Events. Finally, we will hear from Beverly Kuhn with the Texas Transportation Institute on the Application of ATM during Weather Events in Europe.

Now I’d like to introduce Dr. Roemer Alfelor. He is a member of the FHWA Road Weather Management Team and has been leading the Weather-Responsive Traffic Management program area for the past several years. He has been with FHWA for 11 years, the first 3 years in the Office of Infrastructure, before moving to the Office of Operations. He holds a B.S., M.S., and Ph.D. in civil engineering.

Roemer Alfelor

Good morning. I am glad you are able to participate in this webinar. First, I want to thank NTOC for hosting this and Jocelyn setting this up and moderating it. I also want to thank Deepak and Kevin for organizing this webinar and identifying the speakers. Also, I want to thank the speakers we will have for this webinar. As you can see, you'll hear about active traffic management and weather, both on national and international perspectives. This should be interesting.

I will give an overview of the Weather Responsive Traffic Management program at the FHWA and the activities that we are doing at the FHWA. Weather Responsive Traffic Management is one of the three major program areas in the Road Weather Management Program at FHWA. The other two are Road Weather Observation/Forecasting; some of you may be familiar with Clarus and the road weather information systems (RWIS). The other area is Maintenance Decision Support, including MDSS, or winter maintenance. Our program is wanted to find it through the ITS Joint Program Office at the USDOT. We are also involved in a lot of the activities of the ITS Connected Vehicles program.

Why road weather management? You probably have seen this congestion pie before. It shows you the different sources of congestion or traffic delays on the highway. As even see, bad weather accounts for about 15% of all delays in the highways. Of all the non-recurring delays – those that are not caused by regular traffic delays or bottlenecks – it is about 25% of all delays. It is fairly significant.

Our vision and goals for the Weather Responsive Traffic Management (WRTM) are: get a better understanding of the impacts of weather on traffic flow and operations; develop, promote, and implement various strategies and tools to mitigate those impacts; make it so transportation agencies are able to use current forecasts of weather and traffic conditions to manage traffic flow and highway operations; have motorists receive and respond to road weather and traffic information so they can make the right travel decisions; and we want the weather impacts to be incorporated into the existing traffic analysis and engineering models.

This shows the overall framework for WRTM in FHWA. At the core of the framework are the different traffic management strategies agencies can use to mitigate weather impacts. The boxes in light green show what the elements are that influence those strategies. I am going to talk about this framework quickly. First, there are three types of WRTM strategies. Advisory strategies are those that warn or inform travelers of prevailing or future conditions. Control strategies are those that regulate or optimize the traffic flow due to weather events, and they also include active traffic management strategies. Finally, treatment strategies are those that remove obstructions from the roads due to weather events. Last year, we completed and published a report entitled Developments in WRTM Strategies. We conducted a state of the practice: what strategies that are being used by different agencies. We also developed a number of concepts of operations for improved WRTM strategies. We will provide you the link to that report in the chat box.

Now I am going to talk about the different elements that influence what strategies are implemented and their performance. First, in terms of traffic analysis and modeling, we looked at how weather affects traffic flow in terms of volume, speed, and capacity. We did some

empirical studies on weather and traffic. We also looked at the impact of weather on traffic at the microscopic level in terms of car following, gap acceptance, and lane changing behavior. Currently, we are testing and evaluating a weather responsive traffic estimation and prediction system (TrEPS). It is basically it is a dynamic traffic assignment model. We are testing and evaluating them in three cities: New York, Salt Lake City, and Chicago. We will also be developing weather modules in existing traffic analysis tools. We are involved in the SHRP2 project, “Incorporating Non-Recurring Congestion into the Highway Capacity Manual,” and we are in the process of developing Guidelines for Variable Speed Limits during Weather Events. In the area of traffic/weather data collection and integration, we developed a TMC self-evaluation and planning guide, our TMC web integration self-evaluation and planning guide. We have worked with a number of TMCs in Sacramento, Kansas City, Colorado Springs, Wyoming, and Louisiana in conducting a self-evaluation, developing a web integration plan, and implementing those plans. We are also looking at different sources for data to support weather responsive traffic management analysis, both static and mobile data. We have looked at the availability and quality of road weather information for use in WRTM studies. Finally, we are taking advantage of the Clarus initiative and the Connected Vehicles program in terms of the types of weather and traffic data that we can use for traffic analysis

We are also looking at how weather affects traveler behavior, or human factors. Particularly, we are looking at traveler requirements for road weather information during weather events. We are currently testing and evaluating a design guide that we developed for road weather messaging. We are working with several locations around the country in testing and evaluating the design guide. Recently, we completed a Clarus regional demonstration on enhanced road whether content for traveler advisories.

Finally, to evaluate the performance of the weather responsive traffic management strategies and the agencies that are implementing them, we are looking at various measures or indicators that we can use for performance evaluation and monitoring. In FHWA, we have the operations efficiency index for road weather management, looking at WRTM deployment in the 40 largest metropolitan areas. We are also looking at the travel time reliability measures from SHRP2 that are relevant to road weather. We will be monitoring the compliance by states to the 1201 Real-Time Traveler Information System rules. We will be using a lot of the common performance measures that are used to evaluate the weather responsive traffic management strategies. Finally, in the area of active traffic demand management, under the Office of Operations, we are developing a highway capacity manual performance analysis tool of operational improvements. Last October, in Portland, we conducted the first national workshop on weather responsive traffic management. We conducted it jointly with the TMC Pooled Fund program; 26 States, two cities, and one Turnpike Authority, were represented at that meeting. The goal of the workshop was to discuss existing practices, identify needs, and create a stakeholder group. It was very successful. We received very good feedback from the participants. We plan to conduct a workshop every two years.

The workshop helped us to identify some of the things we need to do in the area of weather responsive traffic management in the future. Here are some of the recommendations we gathered from workshop. This will define the direction we will take in the next few years. I just mentioned a few of them: implement the concept of operations that we developed in the project I

described earlier, as well as coordinate with the other operations programs, such as integrated corridor management program, ATDM, and freight. One of the recommendations was to conduct a series of webinars and tech transfer activities. That is exactly why we are doing this webinar series on WRTM, in response to that recommendation. This is the first of a series of webinars that we are planning for this year. The second one has been scheduled for April 19. It will be on using social media during weather events. You will be receiving an invitation for that one soon. The third one is on weather responsive traffic signal management, which will happen sometime in May. The other topics are yet to be determined. Here is my contact information. If you have questions, feel free to give me a call or send me an e-mail. Thank you.

Jocelyn BauerThank you. Our next speaker is Kevin Balke. He is a program manager and research engineer with the Texas Transportation Institute. He has been conducting research in the areas of advanced control systems, intelligent transportation systems, advanced technologies, and system evaluations for over 28 years. Over the last decade, he led or participated in several research studies for the Texas Department of Transportation and FHWA related to weather responsive traffic management. He will be focus on one such project today, Developments in Weather Responsive Traffic Management. Take it away, Kevin.

Kevin Balke I appreciate everybody attending today. It is wet here in Texas, and we are deploying a lot of these weather responsive traffic strategies today as the weather events continue to unfold. What I want to talk about today is, as Roemer mentioned, as part of our workshops we identified areas where there are synergies going on in the transportation arenas. One has to do with merging or talking about incorporating weather response of traffic management into our development of our active traffic management systems. What I want to do today is try to set some of the stage related to that. When we start talking about weather responsive traffic management, what we are really talking about is implementing advisory control and treatment strategies that are either in direct response to or in anticipation of developing a roadway conditions or visibility issues that might be generated from weather. This can be from current conditions that are happening, like a fog situation developing now, or this might be in response to some forecasted weather event, such as continuing fog development or an anticipated heavy storm coming through a particular area. When we talk about weather responsive traffic management, we are really talking about doing are some of the very similar types of things we are considering as active traffic management.

As Roemer mentioned, there are different types of weather responsive traffic strategies. These can be advisory strategies. They can be passive systems or active warning systems or alert systems. There is also a series of speed of management strategies where we are looking at deploying different types of speed restrictions or speed limits depending on the weather conditions. We might also think about deploying some type of vehicle restrictions: restricting vehicles from particular roadways, for example, high profile vehicles in areas that are subject to heavy wind; doing route restrictions, restricting particular vehicles from a particular route; or deploying traction control type devices on a particular route to aid vehicles as they navigate through the system. Some other weather responsive control strategies relate to deploying different types of traffic signal control systems and strategies during different types of weather

events. We are hoping to have a webinar similar to this on using those types of strategies. There are other ones as well: traffic incident management, making sure you have personnel and assets in place to manage during the events. Obviously, during weather events, particularly significant weather events, there is a lot of multi-agency coordination that has to go on. This gives you a general feel of the types and categories of things we consider to be weather responsive traffic management strategy.

Active traffic management is very similar in terms of its concept. We are trying to dynamically manage both of the recurring and nonrecurring congestion that occurs out there on our roadways. With active traffic management, we are really focusing on trying to maintain trip reliability and to maximize the efficiency and effectiveness of each individual facility, at a micro level or an individual facility level. We are looking for things like increases in throughput and safety, or automation of travel decisions so we can quickly and efficiently monitor and change our traffic responses as traffic and travel conditions change through the system. Many of you have probably seen this slide here that lists some of the general strategies that are available and that are commonly considered part of active traffic management. It lists some of the benefits of where these strategies might be deployed. If you look through the list, a lot of the things that we consider active traffic management strategies also fit within weather responsive traffic management strategies.

The other thing that is unique about active traffic management is that it is in its infancy in the United States, and we have a couple speakers that will talk a little bit more about what is actually going on in the active traffic management arena. There is a significant investment being made in the infrastructure to actually do active traffic management. We felt like there is a good opportunity to start using some of this infrastructure to help us do a better job of managing traffic during weather events.

We have been looking for unique opportunities for combining weather events and active traffic management. What are these opportunities? Active traffic management and weather responsive traffic management are both focused on making proactive and dynamic improvements to the transportation infrastructure in order to better manage that infrastructure as a whole. We’re trying to be as responsive as we can at the micro level, both in terms of the facilities we’re working on and looking at short-term improvements and short-term timeframes for making those improvements. The focus is on the traveler itself – improving traveler reliability and improving safety. The infrastructure needs that occur for both doing active traffic management and weather responsive traffic management are similar. One of the unique challenges that comes up with trying to tie active traffic management and whether responsive traffic management is the weather itself. Weather is very unpredictable, even at the micro level looking at individual facilities: it can rain in one spot and not in another, even on the same facility. How do you incorporate all these potential differences in weather into an active traffic management strategy? Then, managing those expectations: what do drivers expect to happen in terms of active traffic management and weather responsive traffic management. There are also a couple of unknowns. As we move forward trying to merge active traffic management and weather responsive traffic management together, we need to think how these strategies we’re deploying can impact the driving task itself. If anybody has driven during a severe weather event, you know it can be a very stressful, and the driving task gets very focused. If we use alternate strategies and devices

and multiple devices and repeating messages, how does it all fit into the driving task? Another unknown is the connected vehicles side of things and what happens today with the active traffic management and how we incorporate that active traffic management into the connected vehicle environment. These are all things, as we begin to look at these linkages, that I think we need to pay attention to.

We (TTI, Battelle Institute, and McFarland Management Group) did a study for FHWA on Developments in Weather Responsive Traffic Management. In the report, we talked about and reviewed some of the WRTM strategies, we talked about how to improve them, and we came up with some concept of operations on how to improve in doing weather responsive traffic management. Active traffic management is just one of the strategies that we developed some concept of operations for. I wanted to briefly touch upon those as we move forward. We looked at, as part of our concept of operations, inside this document that we put together. You'll find a concept of operations that talks about using ATM-type applications for doing weather responsive traffic management, speed management type of controls. We go through the process of laying out how you might go about doing speed advisories for weather responsive traffic management or doing things like variable speed limits or speed restrictions or a particular type of vehicle – for example, setting a different speed limit for a truck during a high wind event. In the concept of operations document, we laid out how you might go about doing that in terms of deploying those types of strategies for a weather event.

Similarly, we looked at how you might use the ATM infrastructure to do things from a weather responsive standpoint. What would happen if you had a lane closure due to ponding on a roadway, or you might want to do something such as separate trucks from the mainstream vehicles during a fog event? How might you go about accomplishing that using the ATM strategies during a weather event?

The concept of operations also talks about how you might deploy and use the infrastructure to implement vehicle restrictions on a roadway. Let’s say, for example, you want to prohibit trucks of a certain size or height on a facility that is equipped with ATM during a high wind event, or, again, restricting vehicles to individual lanes to separate trucks from passenger vehicles during a fog event. We also looked at what was needed to link weather events to ATM strategies. The important thing to remember is we are not trying to replicate or do things in addition to ATM, we are trying to overlay and use those systems. One of the things we talked about in our concept is how you go about the process of implementing and including weather information into the decision-making process? You utilize current weather information and forecasted weather information. These are all needs that need to happen in terms of making a decision from a weather responsive standpoint. You also do this through some of the modeling process; come up with estimates of the impacts of different types of weather events on the capacity of a facility, and get those down to the facility level. Also, how a weather event changes some of the traffic patterns and traveler behaviors that are associated with it. These are all things that need to be incorporated into the ATM decision-making process.

We foresee that this is a layer that lays on top of your typical ATM decision-making process, where all the elements are integrated together in a decision-making process, a weather monitoring decision support system that brings together this information and provides input into

the decision-making process on the ATM side. This decision support system looks at collecting, integrating, and synthesizing all the information that is out there and generating facility-oriented surface level area predictions of what is happening with the weather. It might also include some tools for assessing and predicting the impacts of weather on operations, specifically knowing when it will happen, where it happens, how long it happens, and to what extent it will have an impact on traffic operations. It incorporates a lot of the estimation and predictions tools that Roemer has been talking about and developing through some of his other work. This would tie in with the ATM decision support system that looks at both measured conditions and forecasted conditions so that the decision making process, when you are looking at deploying an ATM strategy, is more focused on both the spatial and temporal types of decisions that must be made. It also provides constant feedback and monitoring of the strategies that are deployed. In summary, as we begin thinking about and deploying our ATM systems and about what can we do from a weather standpoint, we ought to look at these synergies that might exist between the two different elements. They are not far apart in terms of what we’re trying to accomplish and do in both WRTM and ATM. We have three speakers who will show you how some of the agencies are beginning to incorporate the use of different types of ATM strategies to do weather responsive traffic control. The speakers will provide you with some good information on the details of those systems. With that, Jocelyn, I completed setting the stage for our speakers and wanted to turn it back over to you.

Jocelyn BauerThank you, Kevin. Our next speaker will be Vince Garcia. He is with the Wyoming Department of Transportation. He has worked as an engineer there for the last 28 years in the capacity of a structural engineer, a road design engineer, Wyoming DOT’s IT Program Manager, and, since 2003, he has worked at the GIS, ITS program manager. Vince, would you like to take it away?

Vince GarciaThanks. The first thing I will speak to is the real application of variable speed limits. In Wyoming, we have Interstate 80 that traverses the southern part of the state. It includes about 400 miles of interstate, and about 35% of that we now have on variable speed limits. Fifty percent of the traffic is dedicated to commercial vehicle, which is kind of unusual for an interstate. We have various sections that have different AADTs, but we typically have about 11,000 vehicles per day. The peak elevation along I-80 is about 8,640 feet. We see some unique weather events, including strong winds, storms, heavy snow, icy conditions, visibility problems, and drifting of snow. In the past, we’ve had a number of high-profile crashes resulting in fatalities, as well as frequent closures, which can last for extended periods, and crash rates along I-80 have been high compared to other regions of the state. In 2008, legislation was passed the granted us the authority to set speed limits based on a vehicle or weather emergency. Specifically, it allowed us to use variable speed limit signs.

These are the four sections we have in general. The two highlighted in green are the first two sections, and the two highlighted in red are the last two sections. A typical variable speed limit segment will be instrumented with speed sensing, with the signs themselves, and with a number of road weather information systems. You can see the density on a 52 mile corridor where we have a number of different technologies. The sign technologies we've used to have been scrolling film and amber LED. I should point out that we’ve had issues with the scrolling film.

They are very mechanical in nature and they are difficult to maintain. They are also less visible. You will see that later in the presentation. Our variable speed limit approach has been to provide as accurate and timely regulatory information to the travelers as possible. What we really want them to do is to go the same speed. We’ve located variable speed limits typically at interchanges so that people getting on the system will know the speed limit, and we’ve located them on both the median and the shoulder for visibility. If you're passing a truck, you'll see it on the median side, as opposed to having it blocked on the shoulder side.

Our approach is to have people go as tightly as possible at the same speed. Looking at this speed distribution curve, if we had 100 vehicles traveling at 55 miles per hour and 100 traveling at 95 miles per hour, then that is a dangerous situation, especially given potential visibility issues. What we really want to have happen is for the speed distribution to tighten up. Secondly, what we want to have happen is we want to adjust the speed so it’s appropriate for weather conditions. I am going to speak little bit toward the protocol we use in Wyoming. It involves a lot of cooperation between our troopers, maintenance personnel, and TMC personnel, any one of which can adjust the speed limits. When a request is made, the TMC makes the adjustments. It is called out on the radio so everyone is informed. It takes just a minute or two for the TMC to receive the request, reduce the speed, and use your information systems to let the public know about it. Depending on the conditions, we may adjust speeds several times per hour. It could vary based on section and it could vary by number of times per hour.

In the absence of a visual inspection, I want to go through the protocol we use there. If we do not have patrol or maintenance personnel on the roadway, the TMC has approval to adjust the speeds. They will use both the road weather information system as will the speed sensors to make those adjustments. Loosely, what we use are the AASTHO Stopping Sight Distances. For example, if we see a visibility issue where the pavement issue is wet and we have a visibility of less than 500 feet, we set the speed limit to 50 miles per hour, and the TMC will make that adjustment. We are building more detailed algorithm with the University of Wyoming. That will be coming to us very soon.

Giving you a histogram of what has happened in the past, if you take a look at the ideal histogram, this is based on real data. This is based on speed sensors that were in during a storm in December 2009. What we found is, when the situation was dry and clear, we had a speed distribution that was an average of 75 miles per hour. When the storm began to impact, we saw that the speed distribution flattened considerably, with vehicles traveling as slow as 32 miles per hour and as fast as about 80 miles per hour, so a diver speed distribution. As we put the speed limit into effect, what we see is the speed distribution tightening considerable and getting the effect we look for – having people go the same speed. What we found is that we’ve historically held onto the speeds too long, and the speed distribution starts flattening out again. We need to work on that.

Here are a couple representative scenarios of what happens. A patrol trooper might be on I-80 working a crash that might’ve been due to icy conditions, but the icy conditions might not be out there anymore. What the trooper will do is, if the storm is passing, he will contact the TMC on the radio and request a speed reduction. The TMC operator will make the reduction and call out on the radio scanned by both patrol and maintenance to make sure everybody is aware of the

speed adjustment. The TMC operator then turns to their systems and make sure the web, phone, texts, and email inform the public. Our intent is to couple these in the long-run so that when a speed adjustment is made, it just our systems directly.

It’s a very similar scenario for a plow operator. If they see visibility issues due to fog, for example, the plow operator can contact the TMC and ask for a pace speed. We approximate the pace speed to be the average speed plus 5 miles per hour, which has proved to be very effective for us. We grab speeds bins of speeds based on our speed sensors and use the average, plus 5 miles per hour. The TMC operator reduces the speed, calls it in on the radio channel, and informs the public. In the absence of patrol or maintenance, where the TMC, based on sensors, is seeing a visibility of less than 600 feet and dry pavement in a fog situation, for example, the TMC operator is authorized to make the speed adjustment. They’ll use the table we created for them, call it out on the radio, and, again, make sure they put it into the system and inform the public. One of the issues we have had is that there appears to be one section that has caused us a little more trouble than others. It is the section between Cheyenne and Laramie. Speeds are not being raised as soon as they should we. I think it is a cultural issue. There are maintenance people that are reluctant to make speed adjustments. It is one of our newest sections, so it probably hasn't filtered down to the plow operators that they have the ability to make these adjustments. Recently, we had a meeting with our executive staff, the district engineer in the representative district, and patrol personnel, and tried to make sure that they understand that everybody has to take ownership of the speed limit, that everybody has to participate, and that cooperation between patrol, maintenance, and the TMC is a must. Their participation is critical.

We’ve deployed both the scrolling film signs, a non-lit sign very much like a static speed limit, and the amber LED VSL. We’ve done a recent survey that indicates that the amber VSL is more visible when we have visibility issues. What I’m showing here is that in a morning condition with fog, the amber VSL is visible from 830 feet. The scrolling film sign probably is visible 400 feet; we do have a beacon at the top to draw attention to it, but the visibility is still reduced. The white VSL sign is visible at 615 feet. In dry, clear and sunny conditions, the amber VSL is still the most visible. On a dark, clear type of day, the white VSL stood out a bit. The amber VSL and static signs, both black on white, had approximately the same visibility. With that, those are the things that I wanted to bring to the table. I guess we will take questions based on the chat box and I will address those later in the discussion. Thank you very much.

Jocelyn BauerThank you, Vince. Our next speaker will be Brian Kary from the Minnesota Department of Transportation. He has been working with MnDOT since 1999. His work with the department has included working in freeway operations, traffic analysis, and incident management. In his current position, he oversees the daily operations of the Regional Transportation Management Center, which includes the MnPASS express lanes system, ramp metering, and the newly deployed active traffic management system. Brian, feel free to start when you are ready.

Brian KaryI'll start with an overview of what MnDOT is currently doing with ATM and what our plans are. I will let you know how we are looking at improving the system to handle weather events. This first slide shows a picture of our active traffic management system on I-35W. This was done as

part of our Urban Partnership Agreement (UPA) project, where we converted a carpool lane into a MnPASS HOT lane. This gives you an idea of what the sign structures look like. We have lane control over every lane, and the signage is based every half mile. If you look in the background, you can see another set of signs downstream. This covers about a 16 mile corridor, I-35W, connecting the southern suburbs with downtown Minneapolis.

We have about 190 intelligent lane control signals (ILCS). The ILCS are full color matrix 4 foot by 5 boot signs, and we’re primarily using them for incident management, speed harmonization, and priced dynamic shoulder lane operation. Here are some of the sign options that we are using for incident management or if there is road work or maintenance work on the corridor. We are using a lot of the standard MUTCD symbols, although there are a couple of exceptions there. The flashing yellow arrow is something that has been in Minnesota MUTCD for over 20 years. We’ve been using that outside of our Lowry Hill Tunnel and to display cautions we’re continuing to use that on the I-35W corridor. For the merge symbol, we are doing a request to experiment now. The University of Minnesota is doing human factor studies to assist us with that. We’re also doing the variable speed limits and using a white diamond to display when the HOT lanes are in operation.

Getting into the variable speed limits, they are advisory only. It is amber on a black background. What we are doing is detecting traffic speeds downstream and essentially providing a queue warning type system for that downstream traffic. We will post speeds up to a mile and half downstream depending what the speed differential is downstream. It could be up to three signs as you lead into the congested zone, giving the variable speed limits.

This is another way to represent how those speed limits work. Our goal is to mitigate shockwave propagation from downstream bottleneck by gradually reducing speed levels of incoming traffic flow. The X axis here is the direction of traffic, which is going to the left of the screen. The white dots represent our speed sensors along the corridor, spaced every half-mile, and you can see where there is a speed to drop as you come into a bottleneck. That first location of the queue would be at the purple dot. Rather than having people come in and abruptly slow down, we are trying to ease them into that congested zone, and the yellow represents the variable speed limit signs and how we try to step those folk down as they come into the zone.

This is another graphic to represent the same thing. The red box represents the congested traffic on the corridor and the speed limits that would be stepping down as you come into that congested traffic. As that congested traffic steps back or increases, those signs step back as well. One thing we are doing is that the lowest speed limit that the signs will go to is 30 mph. If speeds at the location of the signs drop below 30 mph, we actually blank out the sign. They wouldn't see another sign until they get out of the congested zone, and then they would go at the normal speed.

Here some of the things we are looking on VSL control during weather, some of the difficulties we have been seeing, and things we want to improve upon. The first is that the VSL system uses a constant deceleration rate to determine the advisory speed limit values. We are sensing what the upstream and downstream traffic speeds are, calculating the difference, and then calculating what the preferred deceleration rate would be so that traffic can make a safe speed adjustment as

they approach the congested zone. What we are noticing is that that would be different depending on what the weather conditions are like. The University of Minnesota, Duluth campus, has been assisting us in developing the algorithm. They did a pilot project for us a number of years ago in a work zone. We built upon that pilot project to develop the algorithm for I-35W, so we are now in phase 2 of that research to identify some of the improvements along the corridor, including improvements for how we deal with potential weather scenarios. The next thing is that, as I mentioned, the lower speed cap is 30mph and the upper cap is 5 mph less than what the posted speed limit is. Much of the corridor is 65 mph, and the other half is 55 mph. This could warrant a 60 mph speed limit in a snow event. We’ve gotten some concern, especially from our maintenance folks, about whether we should display a 60 mph speed limit during a snow event and that people should be traveling much slower than that. Right now, we’ve set a default cap of 50 mph, which puts it in line with the 55 mph zone. We are working through that to try to come up with a better speed limit during those types of situations. We have one weather station on the corridor right now. We have a rain sensor, so we can look at the rate of rainfall as well as snowfall. Using that data, we can make different thresholds for the speed limit system so that we’re not just operating during the more typical, sunny day scenario. That’s still something that’s being developed and researched. Wrapping up on that section, we are developing those new techniques for VSL operations during weather events by utilizing some of the lessons learned that we have experienced since we activated the system. The system was activated in July 2010, so it's been a little over a year and a half that we have been operating the system.

The MnPASS HOT lane is the entire 16 mile stretch on the same length of corridor as the ATM. The last 3 miles of the MnPASS lane actually operate on our dynamic shoulder, what we call our priced dynamic shoulder lane (PDSL), considering it’s a priced lane when it’s open; otherwise, it is a shoulder during off-peak periods. The signage over the lane is that carpools, buses, motorcycles, and MnPASS are allowed to use the shoulder when the green arrow is lit. As I said, the PDSL is the last 3 miles there. You can see the corridor highlighted on the right-hand side of the slide. The green section is the PDSL. The red section is the old carpool lane that was converted to a HOT lane. The blue section is a section of roadway that underwent a full reconstruction to add the carpool lane, which was then opened as a HOT lane on the day of opening. That piece of roadway was under already under construction at the time we won the UPA grant. It was originally intended to be an extension of the carpool lane, but it is now a full HOT lane all the way through the corridor. After the construction was completed, it was anticipated that we would've ended up with a five-lane section leading into a four-lane section. We would've lost that MnPASS lane about 3 miles shy of downtown. The PDSL allowed us to get that MnPASS lane all the way into downtown. Otherwise, we would've ended up with a major bottleneck shy of the downtown area and not provided the full of efficiency and benefit of the MnPASS lane. We were able to do that for a tenth of the cost of what it would've cost to reconstruct the last 3 miles. The last 3 miles has been estimated to cost anywhere from $350-$500 million and was planned for another 15 years. We were able to do it for about $35 million, including the technology and the roadway infrastructure improvements, so we were able to do that at a tenth of the cost and years ahead of schedule. The PDSL shows how you can utilize technology to maximize the roadway that we have out there currently.

Here is another artist drawing of how the signage looks and how the lane assignments were established on the corridor within the PDSL section. We basically ended up pushing the mainline

to the right, so we lost the full shoulder on the right and gained more shoulder on the left, because we are operating the PDSL on the left-hand side. We have effectively gone through a 3-foot outside shoulder, just a reactionary, so we do have emergency pull-offs on the corridor. On the left-hand inside shoulder, we made enough room for the express lane as well as the 2-foot buffer between the MnPASS and general purpose lane.

Here are some pictures of the transition area. It is unique in that the HOT lane actually continues up from the south for the last 12 or so miles. We are requiring people to merge left out of that lane, so it ends up being a left-hand lane drop when the lane is closed. We start with the merge sign, start showing that the MnPASS lane is closed by showing there isn’t a rate assigned. If you look into the distance, you'll see a red X. That is the actual start of the PDSL. We are telling people they can no longer drive there. This is a close-up of that lane. That is where the red X is. You'll also notice the in-road lighting we have there that is a yellow taper that lights up when the lane is closed. When the lane is open, that yellow taper turns into it a white skip line to show the lane is open. We have had problems with that. It survived the first winter, and almost made it through the second winter. It wasn't an issue with the plows, it was more of an issue with corrosion in that the salt material got into the lights and caused corrosion at the connections, and the lights ended up going out. We never had any of them get damaged or plucked up by snow plows. We are still evaluating whether we would go back and reinvest to try to get those back up and running or if we would just continue to operate it with just the overhead signage. Here's how it would look with the lane control.

How are we operating that during snow and ice events? One recommendation we got from our maintenance folks was that it would be preferred to keep the PDSL open during snow and ice events. That way traffic can help to spread out the de-icing material; otherwise, we end up with clumps of de-icing material and clumps of ice and snow pack, and it wouldn't us spread that out. That has worked pretty well for us; even though traffic volumes might be lower during those off-peak periods, it’s still worked pretty well.

Finally, one of the last things we are looking at is trying to use ATM during rainfall. In the southern end of the corridor, to get the buffer between the MnPASS lane and the general purpose lane, we had to move the MnPASS lane a couple feet onto the inside shoulder, but that inside shoulder wasn’t originally designed to allow for some ponding during heavy rainfall events. We are looking at ways to utilize the ATM and rainfall sensors to potentially either display a warning message over the inside lane or advisory speed limit over the inside lane to warn people to slow down because there might be an inch or two of ponding water on that lane as they are driving. With that, I’ll remind people that we will be taking questions in the chat box and will answer those at the end.

Jocelyn BauerThank you, Brian. Our next speaker is Dr. Beverly Kuhn. She is a division head and senior research engineer with the Texas Transportation Institute, a part of the Texas A&M University system, where she specializes in the conduct and delivery of research results in the area of transportation operations, especially in the areas of active transportation demand management, active traffic management, and managed lanes. Beverly, feel free to start when you're ready.

Beverly KuhnThanks a lot. What I am going to speak to you about in the next 15 minutes is the application of ATM strategies during weather events in Europe and how the European countries are addressing or taking advantage of their infrastructure to address challenges during weather events. Here’s just a bit of a recap. Technically, the definition of ATM is to dynamically manage recurrent and non-recurrent congestion based on prevailing traffic conditions. The idea is to maximize the effectiveness and efficiency of the facility, to increase throughput and safety, and to use integrated systems in a proactive manner and in ways in which they can automatically be deployed to be proactive and stay ahead of the congestion whenever possible.

The four strategies that the Europeans have deployed for a number of years that I am going to address are listed here. You have heard from my colleagues in Wyoming and Minnesota on how the US is starting to deploy these strategies. What I will address is how it has taken place in Europe and how they deal with weather related conditions with these strategies.

The first one is shoulder use. You will see here on the slide, three deployments. This operational strategy has been in place for at least a decade in Europe, if not more. It is common. Probably the most recent deployment has been in England, starting in 2006. One thing that is not evident in these is it is frequently deployed in conjunction with variable speed limits or speed harmonization. The combination of those two strategies is what gives ATM its power, especially with respect to weather related events.

The temporary shoulder use, you might've heard it called hard shoulder running. The idea is to open the shoulder to traffic to temporarily increase capacity. It is typically done when congestion starts to build and, as Brian said, you want to reduce the likelihood of a shockwave back through the traffic stream. You want to take control of how the traffic behaves in a congested condition. Again, it’s often combined with variable speed limits, which might also be heard as speed harmonization. Variable speed limit is more commonly used in the US. The predominant location of these strategies is in an urban and suburban environment where there is heavy congestion. That is little different than the Wyoming approach, which is a rural application, but I think the advantages are very similar. How can we take advantage of the traffic to control what is going on? With shoulder use, the typical philosophy is that having that shoulder open, combined with the variable speed limits, you can take advantage of that additional lane and the control signage that goes along with it to control the flow during a weather event. If the shoulder is not already open, agencies will frequently open that shoulder when the congestion starts to build as long as they know that the shoulder has been plowed, if there is not any water ponding on it, if it has been salted, or if its visible by the operators during fog. These corridors are implemented with cameras where the TMC operators are able to see whether or not the shoulder is clear of disabled vehicles, debris, etc., has it been plowed and salted, and can they see the shoulder during the fog event. This is how opening the shoulder and taking advantage of it helps reduce some of the impact of the weather related events.

The next strategy is variable speed limits or speed harmonization. Here are a couple of deployments that are prevalent in the US. The one in Denmark was actually during a major work zone on a ring road around Copenhagen. It is currently not operational. It may still be operational on the general purpose lanes, but that example is during a work zone event. It

became extremely popular with the operational agency because during its deployment, when there's a lot of congestion and there are narrow lanes in the work zone, they did not see an accident. They attributed that to the speed harmonization that they deployed. That illustrates how controlling the speeds due to either congestion or some non-recurring event can help control the traffic stream and reduce the likelihood of crashes, both primary and secondary. The idea of the speed harmonization is to control the speeds as congestion starts to build either as a result of an event, an incident, or just recurrent congestion. The weather realm, when congestion starts to build because of a weather event, speed harmonization can be deployed when that weather starts to create sudden disturbances in the traffic flow. It controls the traffic and reduces the likelihood of the shockwave rippling back through the traffic stream. It prepares drivers for the conditions ahead; it provides the quickest warning to motorists that conditions ahead are changing, and it promotes safer driving in these adverse conditions by controlling the traffic and making sure that speed of distribution is tight, as Wyoming likes to see during those congested events so that you don't have that speed differential and abnormal or irregular driver behavior, so that they reduce the likelihood of crashes.

Queue warning is an operational strategy that is deployed very frequently in conjunction with some of these other strategies, but specific to the building of a queue somewhere on the facility. The idea is that a queue is building either because of recurrent congestion where you have a bottleneck or some event, such as a weather event or an incident, and the driver is specifically alerted to the fact that there is a queue ahead. In the illustrations in Germany and Norway, the symbol with the diamond is the European symbol for a queue up ahead. It indicates that a queue is building. It is often complimented with flashing beacons. In the Netherlands, the lights on each side of the speed limits flash. You don’t see the queue symbol, but the lights flashing around the speed limits indicate that there is a queue ahead and the driver needs to be aware of what is happening. The idea is, it is based on the dynamic traffic detection, very similar to the speed harmonization, and it uses those warning lights and flashing lights to tell the driver something is happening ahead and to pay attention. It is often used with speed harmonization, and when queues start to build because of weather conditions, it's even more beneficial because the drivers know. Similar to what Kevin said, you might have some area with rain, and 3 miles up the road there is no rain. Wherever that weather event starts to occur, the operators use that queue warning to alert the drivers. Again, it increases their awareness of oncoming risks. When combined with speed harmonization, it promotes more uniform behavior and safer driving habits during those adverse conditions that are creating disturbances in the travel flow.

The final strategy that is deployed overseas that helps with weather related conditions is dynamic rerouting and traveler information. The illustrations here are locations where agencies aggressively used dynamic rerouting to help drivers address changes in what’s happening in the traffic flow, whether there’s an incident ahead, and how they can be rerouted through the general area to their destination. These signs, they change specifically. If the shoulder is open, then signs will appear over the shoulder so that it looks like a general traffic lane, and guidance information is provided when that shoulder is activated. The concept is to provide that information in a dynamic matter. There are two methods that are typically used. The operator will either suggest an alternate/secondary route and remove the original route from the signage, or they will leave the original route and provide the secondary route information as well. In weather applications, if congestion is building up ahead, there are incidents occurring, the TMC operators can provide

guidance in response to these incidents and conditions to let the drivers know there are alternate routes. One important issue or caveat to mention is that the TMC operators operate this strategy in close concert with the local agencies that are operating the adjacent roadway network. They want to make sure that if an alternate route is displayed to the motorway facility users, they aren't leading them down a road that has and major work zone. They work closely with those agencies to make sure that the information they provide is not going to get the drivers into worse conditions than they were in on the motorway. I think that is really important to take note of. There might be an incident occurring on an adjacent system, and we want that to make sure those strategies are complementary and do not create more problems.

A lot of this information is documented in these two international scans. The web addresses on the FHWA website are located here. The one on the left was a more holistic approach; an investigation of the strategies and some of the weather related information is included. The scan on the right in 2010 focuses primarily on the geometric design aspects of these strategies and how they dealt with shoulder width and the design of the signage, and those types of issues related to the design of the strategies and how they were implemented. There's a lot of good information in there and I encourage you to take a look at those if you’ve not do so already in the last couple of years. With that, I will stop, and we are ready for questions.

Jocelyn BauerThanks to many of the presenters who have been answering questions in the chat area as we have been going along. I have been working to keep track of which questions have been answered and which have not, but, I may have missed some of them. If you’ve already answered a question, feel free to let me know. For now, I'll start out with the questions that I believe have not yet been answered. One is from TxDOT for Brian. Where can I find detailed drawings and specifications of ILCS?

Brian KaryIf you want to shoot me an e-mail, you can. We don't have anything posted. I can provide my contact information: [email protected].

Jocelyn BauerAnother question for you: does the HOT lane have to abide by the variable speed limit on the other lanes, or does it operate independently?

Brian KaryIt operates independently. We had a lot of discussions on that as to what we would post over that lane. That is why we ended up using the white diamond. We are guaranteeing a 55 mph trip in that lane, so we didn't want to put a speed limit less than that. Also, our speed limit folks didn’t want to have a separate speed limit on that either, we wanted it to be consistent across all lanes. We tried to appease all the different users and stakeholders, so we ended up using that white diamond that is displayed during peak hours when the MnPASS lane is in effect. Otherwise, outside of the MnPASS lane, when it operates as a general purpose, it will be incorporated into the other variable speed limit signs. That would only be in an incident scenario.

Jocelyn Bauer

We have a question coming from New Jersey. What challenges were experienced to have these projects incorporated into the capital programs within your agencies? Have performance evaluations been released to support these types of projects? This one is open to any of the speakers.

Vince GarciaThe first segment of variable speed limits in Wyoming was based under a research project, and the remaining segments were based on that research. We will be putting in a fifth segment on a two-lane highway in October, based on that research project.

Brian KaryTo try to answer the question for Minnesota, I think ITS has become fairly well accepted in Minnesota, so it hasn’t been too difficult. The biggest issue we have with this project was that because of the UPA and the grants, there was some difficulty in changing some of the capital program investments to meet the needs for that grant and providing the local match and putting I-35W at a higher priority than other corridors. Otherwise, the idea of ITS and traffic management was pretty well accepted. As far as evaluations go, they are still ongoing. We have the University of Minnesota assisting us with that. There are the larger UPA evaluations that are being done as well.

Jocelyn BauerIn the European example, is speed limit regulatory or advisory?

Beverly Kuhn The examples that we showed were regulatory. The signs are a little different. There's some where the speed limit has a red circle around it some where it doesn’t, but they are all regulatory. I know two different people asked whether there has been a study of advisory versus regulatory and if one is more effective than the other. I'm not familiar with any study, at least that's been done domestically. At least overseas, I think all of their signs are regulatory, so it is a non-issue. I know that Washington has some regulatory signs and Minneapolis has those advisory signs. There hasn't been a side-by-side comparison yet, as far as I know.

Jocelyn BauerDid MnDOT ever consider making the last 3 miles into downtown a general purpose language as opposed to a priced dynamic shoulder lane?

Brian KaryNot exactly a general purpose lane, but it would be a HOT lane, just extending it in, restriping the road and not having shoulders. What we wanted to do is have that operational test of having a dynamic shoulder lane and seeing how that would operate along the corridor. We have another corridor on I-94 that was actually the detour route when the I-35W bridge collapsed. We utilized the shoulder as a general purpose lane and we are actually now reconstructing that area. We will still have it is a four-lane section within three lanes of pavement where we don’t really have a shoulder, and we’ll have the ATM, but that will be operated as a link 24/7. So, in a way, it will be a slightly different application in that I-35W will have ATM with a dynamic shoulder length

concept, and I-94 will continue to have ATM, but it would just be four static lanes where we can utilize the ATM should there be an incident.

Jocelyn BauerOkay. Can we get an electronic copy of these presentations? Yes. They are available right now, with the exception of Roemer’s, in the file download box just below the chat box. They will also be included, as well as Roemer’s, on the NTOC webinar archive site. That URL was posted into the chat box early on. I will be sending out an e-mail to all of the registered attendees to let them know when that is available. That will be sometime next week.

How is current and forecasted weather information integrated into a weather responsive traffic management system? How are the potential effects on traffic calculated? Is active traffic management applied only when the weather effects are seen in traffic?

Roemer AlfelorAs far as I know, the current weather information that is used for weather responsive traffic management, the idea is to become more proactive in terms of using forecasted weather information. I'm not sure if there are any agencies that are proactive when it comes to weather events and adjusting their traffic operations during weather events, but that is where we want to go. We want agencies to be able to use the forecast information in providing weather information to the travelers. There is a problem with providing forecast information, because it might not be as accurate as we want it to be. That’s one of the reasons why agencies do not provide that information to the travelers. In terms of how the potential effects on traffic are calculated, that is why we are doing the traffic modeling and analysis. We want to be able to understand how weather affects traffic, behavior, both at the microscopic and macroscopic level. That is why we are doing the analysis of existing data on traffic volume, speed, and capacity as they are influenced by different weather conditions. We want to establish the knowledge so we can use it to service inputs in our weather responsive traffic management strategies. The third one, is active traffic management applied only when weather effects are seen in traffic? I don't know. I guess I would have to ask the other presenters about that.

Brian KaryOur active traffic management is more for recurring congestion in the urban areas. We are just starting to look at that and how we can tweak that system and how we operate it for weather events. I would say the answer to that question in short order would be no, we are really using it for all recurring congestion.

Kevin BalkeAvailable through Roemer’s Road Weather Management program, there is a report that talks about how to integrate weather information into TMC operations and there's a self-assessment tool. It is available to allow you to go through an assessment that will help you figure out where you currently are in your weather integration process and what steps you can take to expand upon that. You can find that through the FHWA website.

Vince Garcia

We use our variable speed projects not just for weather related events, but also for crash related events. It is primarily used for weather events, at least 90% of the use is weather related, but it could really be used for virtually anything.

Jocelyn BauerTo address a couple of other questions in the cha box, we are recording this webinar and the recording will be posted on the NTOC webinar archive site within a week, along with a transcript of the question and answer chat box here. You don't need to worry about saving it on your own. The rest of the questions, I see, it looks like they have been addressed. Are there any that you see that have not been addressed for the speakers on the line there that someone would like to talk about?

While you are looking, I will re-ask a question from Brent, and maybe Vince will go into a little more detail on this. His question was: speed limit changes typically require a traffic engineering order approved by the traffic engineer. How do TEO requirements come into play with VSL? Do speed limit changes require traffic engineer approval? This may also be something that others on the panel want to chime in on as well.

Vince GarciaIn Wyoming, I noted the legislation that we use for our variable speed limits, Title 31. In essence, it states that the superintendent has the authority to make speed adjustments. They can do that with an engineering or traffic investigation, or in a weather or incident event. Again, the superintendent is one that is authorized to do so. He has blessed our policy for patrol, maintenance, or TMC operators to make those adjustments.

Brian KarySince ours are advisory, we haven’t really had to deal with that issue. That was talked a lot about. We were debating between advisory and regulatory. It’s part of the reason why we went with advisory only.

Jocelyn BauerRelated to another question, are there any studies that show a reduction in primary or secondary crashes due to VSL?

Vince GarciaI think our research shows that we are seeing a primary reduction. Again, I really want to hesitate to declare victory, but we are seeing very positive results. I should point out, for example, the reduction in commercial vehicles in the first year, we saw about a 50% reduction in crashes involving commercial vehicles. Again, that's about 50% of the traffic on I-80.

Roemer AlfelorI’m not aware of any national study on VSL impacts on crash reduction. There may be something out there, but I am not aware of it.

Beverly Kuhn

The highways agency in UK did a pretty detailed safety analysis after 3 years of using safety data on their M42 deployment in Birmingham. They have seen significant reductions in crashes on the facility. I don't think the report is available. I might be able find a link to it. If I do, I will send it to you so you can access it.

Jocelyn BauerWe will move to wrapping up this webinar. I’d like to give you a little information on the National Transportation Operations Coalition (NTOC). On this slide, you'll see the member organizations of NTOC. We encourage you to go to the NTOC website listed on the following slide to find out more about these organizations. The NTOC website contains information about upcoming webcasts. It also contains a webcast archives page with transcripts and recordings from previous Talking Operations webcasts. As I mentioned, we will have the recording and slides from today's webinar up within a week. It also has two discussion forums, one focusing on high level or strategic issues and the other on ITS deployment and lessons learned. You can also sign up on the website for the NTOC newsletter that is e-mailed out twice monthly.

In conclusion, I would like to say a special thank you to our five speakers and to all of our participants for active participation and all of your great questions. We hope you found this informative, and please enjoy the rest of your day.