 So, I will be talking about traffic load monitoring and the role of way motion in it. Just briefly, the scope of the fact sheet. It basically describes the available way motion technologies, summarizes the bridge traffic load modeling procedures and presents potential of bridge-wim technique to be used for monitoring of bridges. It is as brief as it can be what is in the fact sheet. What I'm going to talk about is first a very brief illustration about the traffic loading in Europe, then how we perform measurements of traffic load monitoring, then the great part will be on way motion and specifically on bridge-way motion, how we use swim data on bridges and of course some conclusions. So just to illustrate the importance of traffic loading, the first fact is that freight traffic will continue to grow regardless of the discussions about the mode shifts and political will about mode shifts the road traffic will grow. So we will face this problem also in the future. I will point out the international haulage, which is a big part of the freight traffic, which has slightly decreased in the EU-15, if EU-15 are the old member states. But contrary to that, this kind of traffic dramatically decreased in the new member states and that's why we have this considerable grow. I have put down here two tables, the first one presenting the five most trafficked or most transited countries in Europe and five least transited countries in Europe and you can see that some countries like Lithuania, Slovenia, Luxembourg, Slovakia, Latvia, there even 90% of all the freight transport is just in transit, which is of course a big burden for infrastructure. With respect to bridge loading, the problem of course is that we have bridges which are very old. In Europe, we did some studies in a few projects, around 60% of the bridges are over 50 years old. There is usually a big peak somewhere, this is an example for Slovenia, other countries have this peak maybe in the 80s, in the 70s, in the 90s, but the fact is that the great majority of bridges is old. This means that they were designed by old codes which are not appropriate for today's traffic. This is an example again for Slovenia, we have used eight codes, eight different codes in the last 110 years. So which means that this is not, many bridges, all those red ones are not designed for today's traffic. Of course the problem is that if traffic load increases, safety factors decrease and this is only acceptable to a certain level. So how do we perform measurements of traffic loading? The most common of course are traffic counters which are fine, but they don't provide you any information about the true axle loads. So trucks have to be weighed. The most conventional way is static weighing, then we have all kinds of weigh motion techniques which I will go briefly through. The most modern one is the on-board weighing, where trucks are equipped with weighing devices, but this at least in Europe will not come in the next few years I'd say, maybe in 10 years. Static weighing is the most accurate and most common weighing technology and in most countries is the only way to detect enforcement and to basically analyze the offenders. These devices are certified, they're calibrated and as I said they're very accurate, but they are not primarily used for traffic monitoring. They're in front of factories, quarries, dumps and so on. This is an example from the United States where they do have this kind of weighing stations, but I don't think there are any in Europe. These are just a few different devices, some permanent ones, axle weighers, these are the whole weigh bridge where the whole vehicle goes on it and these are the portable devices which are typically used for enforcement on European roads. Low speed weigh motion is another technology which weighs vehicle traveling at low speeds up to 10 kilometers per hour. They are almost as accurate as static weighing, but not very convenient for collecting traffic data because there are special lanes that have to be constructed for this kind of weighing and again it is quite time consuming to perform this kind of weighing. So we come to the high speed WIM systems or WIM systems. So they measure dynamic axle loads at highway speed and uncontrolled conditions and calculate estimate of their static axle weight. So basically this is not, by definition, this is not weighing, this is an estimation of the weights of the axle loads. It cannot be weighing because for that you would need a static measurement. These systems typically deliver, of course, exact time when the measurement was performed, single and group axle loads, gross vehicle weight, number of axles, length and axle distances, speed, vehicle classification. This is basically for all systems and then there are specific features for individual technologies. They have been introduced in the United States in the late 1950s. There was a big boom of installation in the 1980s when they thought that the cost of those systems would drop to $5,000 at that time which of course never happened. There was intense development in Europe in the 1990s, especially within the cost three to three action at that time. Cost actions were much smaller. I think we were only about 30 people from 15 countries at that time and some other projects afterwards. And today there is high focus on application of WayMotion, not that much on development of technology but on applications of using these data they provide. These are just examples of high-speed WayMotion, some examples. So there are basically two types of technologies, pavement-WayM systems and bridge-WayM systems. One are used, of course, in pavements. The other one is existing bridges. Most common sensing technologies are piezoelectric, piezoquartz, strain gauges, fiber optics and also some new emerging technologies. The pavement-WayM systems are divided into the plate sensors, which are a bit bigger, and strip sensors which are very narrow, basically. The main difference is that with the plate sensors the whole wheel is on the sensor and that's why they are more accurate than strip sensors. But they're also quite aggressive to pavement because you have to dig a big hole into the pavement which is not good from the durability point of view. Strip sensors are smaller. The prevailing technology nowadays is these piezoquartz, which is definitely the most popular wheel sensors nowadays. They are typically installed in less than one day and they provide relatively high accuracy on smooth pavements. That's fine, but you may have problems in flexible pavements because they deteriorate over time. And with the deterioration also the quality of the results degrades. Bridge-Wim is an alternative technology. So we use existing bridges to weigh the vehicles. We measure basically strains. They appeared slightly later in the late 1980s, but they are only successful in the last 15 years. They provide exactly the same data as the pavement-Wim systems. In the past, we were always facing the question whether this is only for bridges, no, it's not only for bridges. They provide exactly the same information as the payment systems. The advantages are that they're completely portable. They provide high accuracy. You can imagine that the bridge is very long. The weighing platform is very long compared to the payment systems. They are easy to install. Everything in principle is installed from the underneath. So you don't have to block the traffic when you install or repair the system, which is a big advantage. And as I will show in the next slides, it can provide additional structural information about the bridges that are being instrumented. These advantages are that you need a proper bridge, which is not always available. And for less common bridges, I'm not talking about these simple slab bridges or beam and deck bridges, but for more complicated bridges, you need really a bridge expert to perform these kind of measurements. These are just the two examples. This was the bridge we did in Canada. It's a very short culvert. And you know this one. This is the Pond du Milleau where we did measurements together with the EFSTAR Institute. So this system is now very well spread around. Just a few words about the accuracy of WIM systems. This is determined by the combination of the accuracy and the reliability of the measurements, and it's usually expressed in terms of error is within certain percent for Y percent of measurements. So plus minus 5 percent for 95 percent of measurements. So it's a statistical quantity. Criteria usually vary for single axle loads, axle groups and gross weight. Gross weight has to be the most accurate. Axle loads could be a bit less. And again, this cost tree to three provided the WIM specifications, which are now a widely accepted standard all around the world. Accuracy affected by certain factors. First of all, road condition. You can have a perfect system, but if the road is not smooth, you will get rubbish out of it. So you need a smooth road. Quality of installation is important if you have, which happened actually in a few countries lately, that these systems were installed by non-experts. Again, you cannot expect useful results. Maintenance and calibration as well is very important, and also environmental effects. This is an example of non-calibrated data for temperature. You can see how the average gross weights are moving with temperature up and down, which of course has to be compensated for. A few applications of WIM data for bridges. If we talk about traffic loading, we of course would love to have accurate data, both for design and assessment of bridges. And the important thing here is the traffic data varies a lot from one country to another. This is an example we did in FP6 Project Arches, where you can see that, for example, the Dutch traffic is about 40% higher than traffic in central European countries. So if central European countries and the Netherlands would use the same rules, some would be widely overestimating that the traffic loading. And this is basically, as WIM said yesterday, this is quite close to the Eurocode loading, what the Dutch are experiencing on their roads. Free flowing is important for shorter spans and congestive traffic for longer span that's well known. This is usually modeled with extreme value distribution, so with statistical methods, which are fitted to block maximum daily values or maximum weekly values. And there are also some alternatives, so which especially the long run simulations are nowadays very popular. So you model literally billions of trucks to get the extreme loading in these kind of load models. How can we benefit from Bridge WIM and Bridge Behavior? First of all, this is an example how Bridge WIM actually works. If we have measured response, which is the blue line, and we have here five axels of a five axis semi trailer, we need the influence line to combine the individual axle loads into the calculated response. And the whole idea here is to minimize the difference between the measurement and the model. And for the model, we need as accurate influence line as possible. And here is this link to Bridge Assessment. So this CWIM system, which is the Bridge WIM that we are developing for the last 20 years, we have in the TRIM, which was the FP7 project we developed the procedure, which calculates an influence line for every vehicle that crosses the bridge. And once you average those things and this is the average value of 528 influence lines, in this case, with plus minus one standard deviation, you can quite quickly get what we call the true influence line of the bridge. And in the TRIM project, this was also used to investigate whether this can be used for checking the boundary conditions. So to monitor the boundary conditions and it actually worked pretty well. I'm almost finished. The other important parameter is the load distribution, which is quite often guessed or is modeled. In this case, you can, for all sensors that are at the same time used for weighing, you can add extra sensors if you wish. You can get statistically evaluated load distribution factors. So for each measuring point, you get mean values and standard deviations under the random traffic loading. And the third important parameter that you can get is dynamic loading, which is usually evaluated as dynamic amplification factor, the ratio between the maximum dynamic and static loading. And in the design codes, this is typically done by combining the extremes of both. So the static loading and dynamic loading, which is fine for the design, but when you want to do the assessment, this is not optimal. So what you can do, apart from using the conservative design code rules, you can either model that, which is possible, but quite difficult because it's, again, uncertainty about the parameters that are using is a key factor, or you can do monitoring of it. And again, within the CWIM Weigh Motion System, we have developed a method which evaluates this dynamic amplification factor for every vehicle that crosses the bridge. This is an example for this seven times 25 meter bridge. If we would use the design load rule, we would have to use 1.2 for dynamic amplification. And these are individual vehicles, dynamic amplification factors, where you can clearly see that as you go to the higher strains, higher loadings, this dynamic amplification factor goes down. So basically you have this kind of a reserve in dynamic loading, which is very beneficial when you assess existing bridges. These are just a few examples. Again, for the lighter vehicles, this goes extremely lively to say so. For this heavy special transport, you can hardly see any dynamics in the signal. So this is the illustration of the previous graph. So in conclusion, road traffic will continue to increase and knowing traffic data is a key to reducing uncertainties about bridge loading. And by performing WIM measurements, you can really decrease these uncertainties. It's, we are talking about very concrete figures. We have to be aware that loading is very different from one country to another and even from one road to another. So using a load model, which is not appropriate for certain bridge, maybe over conservative. WIM is a proven technology that provides unbiased information. So if you perform static weighing, you will only measure a small part of the vehicles and they would probably be the most heavy ones. So this is not what you want for traffic modeling. WIM measures everything and thus provides unbiased information. Accuracy of good systems are around plus minus five to plus minus 7% for 95% of the results. They require adequate set up calibration and maintenance. Otherwise, the results are very questionable. Most popular are still the pavement WIM systems, but bridge WIMs are emerging. And another point is that if you perform bridge WIM, this provides some useful parameters to improve structural modeling, influence lines, load distribution factors and dynamic amplification factors. This is a very wide area. So if you are interested in more, I suggest to go on the website of International Society of Way Motion and you will find also descriptions of individual technologies and what they can provide. Thank you. Thank you very much. And now we have the last.