 Thank you. Hello everybody. My name is Dominic. I work at the Faculty of Civil Engineering in Zagreb and I will present to you a case study bridge which I developed along with my supervisor Professor Mandicim Lankovic and with Mr. Znidaric from Slovenian National Building and Civil Engineering Institute. So main purpose of the presentation and the whole case study development is using the initial investment in SHM that will result in extended bridge service life and overall more sustainable bridge management. Type of monitoring data, type of SHM data is bridge weight in motion measurements. Here we can see the sensor for those of you who are not familiar with it. Those are sensors that use the bridge as a weighing scale and they weigh the traffic on the bridge but they also give some structural data. So we can divide the data obtained in two categories. First one is traffic information which gives volume, weight, speed etc. of every vehicle passage and other is structural data. We can obtain realistic influence lines, girder distribution factors and dynamic characteristics of the bridge and those types of data depend on how long are we measuring, how many data do we have. Post processing of monitoring data can result in bridge structural data can be used to improve bridge numerical models and traffic information if we have enough can result in site specific traffic load models. A reason why this bridge is chosen this is the cross section of the bridge so it's small bridge single span of almost 25 meters simply supported and it's superstructure it's consisted of five pre-stressed girders and a monolithic deck that connects them. The reason why it was chosen is because the whole assessment of this bridge was done during my STSM which I done in Ljubljana and my host was Mr. Zniloric. We chose this bridge because we have original design plans, reinforcement drawings and everything else available from the archive and also we done a visual inspection and over two years of constant monitoring data is available for this bridge. I will not go in detail about the assessment we done. All of this is available in STSM report if you want to have a look into it. It was presented in fifth workshop of this cost action in DTU last year and also assessment procedure was presented in Zagreb in March of this year. More detailed assessment and outcomes are available in a paper we published. It's accepted it's not published yet so very detailed procedure can be found there. This is numerical model of the bridge which we developed and then we used data from monitoring to calibrate the numerical model in order to obtain better results. What we did is that we put together a limit state function and we chose a critical failure mode as it's a simply supported bridge. A critical failure mode is bending in the middle of the span and we defined a limit state function. This is resistance to bending and this is load effect as a bending moment and we did it for each girder so we took this limit state function and we evaluated each girder separately. We did it deterministic and probabilistic and for probabilistic assessment every variable in this equation was modeled as a stochastic variable. Here we can see for resistance side of the equation what we used, all the variables and distribution which was chosen. The detailed parameters of standard deviation and mean and characteristic values are available in a paper and information that we had and measure we take them from real life measurements. Another we took from probabilistic model code for suggestion of distribution and we had a mean values so we took probabilistic model code suggestions on how to evaluate standard deviation. So this is resistance side of the equation, this is loading and we did a two different type of the assessment. First one we took traffic load model from Eurocode for designing of new bridges which we then took as our prior value and we also did with traffic load model from obtain with bridge weight and motion measurements and we also included, you will see later, structural data like influence line then dynamic characteristics of the bridge etc. So we did a whole assessment and these are the results. Here we have reliability index beta which was calculated with first order reliability method analysis. For each girder this blue ones are our prior values so they are calculated with traffic load from Eurocode and the green ones are calculated with all the SAGM data that we use. And just from this slide you can see that quantification of SAGM measurements is really obvious. As we can see for example girder two which was most critical because it had a priori value less than 3.8 which is Eurocode suggestions for new bridges and with monitoring data it went over 7. So this bridge maybe is not the best example because maybe it would be better to choose a bridge which has priori results below Eurocode thresholds but we chose it because we had all the data and then we wanted to do a complete assessment of it and define a procedure and then we could apply it on some other bridge which is in much much worse state. So this is what I just said result analysis quantification and its foundation for further analysis of this case study bridge through value of information analysis. So implementation of value of information analysis I have it in couple of slides but these slides are basically data from this diagram as it's pretty big so you cannot read it I will go step by step just a quick overview. So decision maker bridge owner is national road director there are no additional state holders from their perspective main objective is optimization of overall management system and priority ranking for bridge maintenance which can be pretty useful when you have limited funds over time periods. These objects are achieved through normal and steady traffic flow on the bridge and extended bridge service life in general. So we can conclude that additional investments in SHM tools can be justified by fulfilling these objects and we will try to prove that through value of information analysis. As for costs if we invest in SHM we have increase in initial investment but we will try to prove that it would pay off through time. Because closing bridge for traffic its income and reputation loss for owner of course and it can also result in more catastrophic consequences like bridge collapse etc. For boundaries for this SHM system there is no restrictions based on bridge type or bridge dimension so it can be used on variety of bridges but it requires qualified personnel and for installation and for data post processing and additional knowledge for probabilistic or numerical calculation methods. These are events of interest sorry this is our prior value assessment according to codes for new bridges and then we had two different assessments with SHM one is short term one is long term measurements because you cannot always have enough data for for example site specific traffic load model. Consequences can be bridge strengthening which can be unnecessary or necessary appropriate or unsuitable traffic restrictions like weight restrictions on the bridge or it can be bridge collapse etc. But everything results in money loss and reputation loss. These are the main indicators that we observe from measurements realistic influence line, greater distribution factors and dynamic characteristics from structural response and also from traffic data that we can develop traffic load model. Resulting is reliability index. These are alternatives so we can decide based on visual inspection we can assume the state of the bridge and decide would we do assessment with SHM or we can do it without. Other measures which can this result is that we need to redefine the use of the bridge, impose weight restriction or demolition and total replacement of the bridge or strengthening or repairing of the bridge it depends on the assessment results. So this is the flow chart for proposed value of information analysis and then based on training school knowledge and based on yesterday help from Sebastian and Daniel we produce decision tree and this decision tree is based on how it was proposed by Sebastian. So first we have strategies we have three so this could be our B0 assessment without SHM and it's our prior value and then we have for example B1 and B2 strategies assessment with SHM level one which means short term measurements and assessment on level two which mean long time measurements but of course you cannot always if you need to assess bridge right now you cannot expect to let measure for two years that's why we have this intermediate step because for this assessment it's enough to have couple of days of measurements. Now after that we have SHM outcomes for these two branches because this one is assessment without SHM and they can show us is there improvement or no improvement in rigidity ability and what is important is to take into account uncertainties costs of course SHM time restrictions because for this one takes a long time and after that we have actions. So regarding the actions we have only two branches but because of the simplification of the decision tree we should have branches for no repair and then three, four possible repair types so we can do bridge strengthening weight restrictions or total bridge replacement and we also have consequence of these actions which are non-exclusive and no matter which action we choose we will have road closing definitely which will cause traffic jams on other roads and lots of money and lots of reputation in perspective of the users and finally we have system states we define just two bridge safe or bridge not safe and of course if bridge is not safe it has direct and indirect consequences these are the ones we saw earlier but we can also have in worst case human casualties and indirect consequences some of them are that traffic jams on alternate roads and lots of reputation. This is just a sketch first draft of the decision tree and I hope that we will develop it through STSM mission that we are planning. So critical appraisal requirements is that we have complete assessment on each level that's why we chose this bridge. What we need to do now is that we need to define all the costs and try to evaluate the bridge importance on the network level because that's important for indirect consequences which is connected with consequences of bridge closing and possible simplifications we can do 3D bridge numerical model and maybe do a simple one with just 2D grader model and we can assume bridge importance on network this one we can do it based on similar bridges. Those are first 2 simplifications that came to mind of course there can be a lot more. So general conclusion is that we have proven that with this that bridge weight and motion has a big contribution in increasing of bridge reliability now we have to prove that cost and benefit ratio is also good. Advantages of this is that we have everything because it was evaluated during the STSM in this cost action. We have complete multilevel assessment and we have detailed 3D numerical models. And further steps of course what I just said we have to define cost and we have to set up whole value of information analysis and that's it. Thank you.