 This is my case study, I think I know I represented it to be tested in the other context. So, basically I don't know if it's just a brief because it's always good to understand. I represent briefly the owner of this bridge, it's a bridge recording, it's proportional to zero, the music is studied. Many three components of the MRID, then a lot of works of the system, monitoring, the structure, internal performance, and the sort of performance that was already done. And then what does the code, maybe the last part of the application, I think in terms of the iterative, our information analysis support is being used. So, basically, the BDSA group is one of the largest multi-resident Brazilian rules. We have investments not only in Portugal, but also in the United States and in the Netherlands. In Portugal, we have, they have already seen the ASJM as a strategy investments. So, there are some bridges with monitoring systems, many tool, department of monitoring systems in Europe as well. The one that I focus on, I chosen for this case study is the zero bridge, which is the second big, honest vision rule. It has a monitoring system that is provoked, it has data system structure. So, the size of the system is quite complex, it measures different parameters, static dynamic behavior is observed. So, when this is a project that is ongoing, and the reason they want to do it further, we do this in the last amount of procedures for this project. So, basically, briefly, what I have to offer for this case study has the right information. A part of all this information from the monitoring system that we collect in the last 11 years, we have also read about a very detailed model that includes not only the comfort elements, but reinforcements embedded by stress, external tendons, the construction sequence as well, very, very detailed. All possible information that was available is included in the model. So, what we could at least try to, from the point of view of the owner, in the first step, show them that he's in data, it's reliable, and they can rely on, because now they rely more on the predictions from the design of the grid, but we were able to show that for this case, with the model that we showed, with the model. In fact, there was a good agreement between the data and the model. Okay, with the small animations, just to understand that the model was quite detailed with the sequence of the structure and the end of the virtual lifetime, it's expected to be 100 years. So, we have this database, a very incomprehensive, let's say, infinite element model database as a complement to the general database. In terms of comparison of the results, I would say in terms of the long-term performance, like this is the main focus, we could see that either the construction phase, the loading test, and the long-term performance, the results agree very well in terms of prediction against the model. But also, there's something that I have here, and this is also a discussion from yesterday. The system is also able to assess, to be able to do traffic loads, so we can see in the left some measurements from the strain gauges, we can see some bits due to traffic loads. From this, and using a calibration model, we could use the data plus infinite element model to estimate the loads that are affected across the bridge. So, based on statistics, you could fit a distribution, and from this, by using extreme model theory, we can estimate the different, let's say, extreme loads associated with different return periods of loads. This work was done, and we can see the safety level of the structure of the deflections, at the mid-span here, at this span, for example, and we will see, we also have done, for this case, the limits we might exceed. So, just to show as well that we have these two dimensions of the problem covered, not only the long-term performance, but also the short-term performance in terms of load loads. So, okay, it's just a brief reference, but for this case, attack on this as a bottom line, I think that it was very useful yesterday, so this is what was done in Munich, which, again, I repeat, I appreciate the support of Michael Farber, I know some of the ones that a lot of you saw me use, these issues here, but from the last meeting yesterday, what I can, for example, in fact, in terms of, for example, characterization of lag grid functions, we can do it very easily for equipment shrinkage models and traffic load models, because we have these already available, so from the data that we have from the ACGEM system, we can, in fact, update the knowledge that we have from the codes to these site-specific cases, either in the long-term performance because we can also have an instance of different loads. It was also, and from Michael Farber, something related that perhaps we could try to look more in terms of traffic delays or diverting the traffic due to this excessive flex of mental curve. And in this case, because this is a private company, it's not only minimizing costs of maintenance, but they have some tools on the grid, so I need to clarify if this should be a minimization or a maximization, either costs or benefit the profits from the tools. But in terms of performance, the one step that I'd like to go ahead perhaps makes sense not only to include the long-term performance in terms of deflections due to equipment shrinkage and analyze this problem of excessive flexion, but with this work that I have already in traffic loads, also the coupled effect of this baseline, which is the long-term performance, plus the traffic loads, which therefore I can use for services in its states the combinations available and with these likelihood functions we can say with and without ACGM if we gain some benefit and understanding better what is the safety level and or the extended period of life, it's possible more, or less, in terms of indicators, I would say that for this case it is possible to explore the flexion of admittance sections due to this issue of the excessive flexions, but also due to traffic loads, if I consider these in the model, the cracks on concrete, mainly on the support sections which might be a critical issue which we can derive it from this very refined, finite element model and in terms of remedial actions also something that Michael suggested we could also look that you could look more in terms of the first instance as a first approach, perhaps is not to repair the bridge but if we start having a pattern of the flexion that might affect its visibility or the comfort of people traveling we may think about limiting the traffic speed and or the weight of the vehicles that cross the bridge and all these can in terms will be the impact in terms of costs or the benefits of the tools and therefore this is what I got from yesterday, so thank you very much I hope that ties Thank you very much Any questions? Yes I don't think that you mean a situation in which you have the limitation of traffic No, I have already the traffic simulation made Yes, you are going to do this with an American model I have already done You are going to use the result of what you have already done to model a future situation in which the traffic limitation So basically the idea is that as I present here from the data that we collected from this string gauge of the isomer rate this is the wind problem I measure the response of the structure and through the model I can estimate the loads that are present on the bridge I can estimate let's say at least we believe that it is more representative these extreme loads that we are getting for this specific bridge and therefore when I am assessing the safety in this case this was a paper because object 2016 is only focusing on the traffic loads but the idea is that and based on the discussion of yesterday perhaps it becomes easier to formulate this problem of other information not only considering this excessive flight just due to group of string gauge but if you load the traffic load I think the reliability might become easier because you can use also the influence lines that can build an Instagram and therefore the response of the structure against the action of the structure so I think it's it makes me more sense but to you I think and it is like a little function where as you may see based on the data you have yes we have the codes there is some assertive for example it is stated that certain of these models it has a coefficient of variation around 20 to 30% but I have data on the bridge from some specimens so I can characterize the uncertainty for this specific concrete on the bridge so I can update my previous models with the posterior on the Bayesian rule and also you started from a model from a field and then you updated it based on the bridge by a client base for these I didn't done yet with this paper that I did I have the characterization of the traffic loads so now I can took this histogram as a likelihood combined with the prior from the codes because in the codes you have the neural code 1 you have mode model 1, 2, 3, 4 I can understand better because Brisa has tools which is quite interesting because they record all the vehicles across the bridge therefore we can and they don't wait don't have the weight of the vehicles but they have the class on the vehicles so it's 2-axle, 3-axle so I make also update load models from neural code for this specific use so therefore I have a prior, a likelihood and I can have a posterior for the traffic simulation well, this is Bapito Mayer since yesterday so don't ask me too much because you can ask but I may become a little bit but one of the next steps and I'll continue with this study so basically, there is a lot of work in traffic load analysis it's one of the things that I'm also doing some research on okay but from my knowledge there is no such work that combines these reflections due to long-term performance and the traffic loads and from a serviceability point of view if you look to the load combinations it makes no sense because the load combination for serviceability is your counter-contribution of the permanent loads plus live loads so the permanent loads in this case people assume that it is made into dead loads and press pressure showing by dead loads but principally so your decision on the measurement is whether you would use all your existing installations or yes, so if I would use this information exactly and then the risk you analyze is the risk you calculate compared based on the design equations from the Eurocode and you compare the risk reduction you gain when you use your refined model so basically the other thing and the actions the actions should be the traffic loads no no no a lot of actions the actions you could do with my steps would you ever have the traffic loads so the first thing that I thought that might also be mentioned that you should start with the perfect information without the pain but I think also to build up this from the bottom to the up the first action should be I think it makes sense if it starts having some discomfort in the channel to reduce traffic or if the safety is put in more because we are starting also to see some cracks perhaps also reduce the weight of the vehicles if this goes to another level that is put in cause not the safety but the serviceability levels are not satisfied that might produce cracks in the long term might produce corrosion and so forth perhaps in that case, in that last case you could try to repair the bridge but I don't know yet because we discussed ideas of that perhaps stress is not always the best solution because I include the more creep more the fractures but there are other people that are studying how to do this which is quite the creep specific problem I have to ask you a quick question I think that is really interesting what you said at the very end but looking at traffic loads and their influence on creeper trinkets that is no relation so are you talking about equivalent quasi static load related to the traffic load so saying that if it is a heavily traffic structure based on the frequency of vehicles and the weight profile of vehicles that there is an influence on creep due to the traffic load that would be really interesting if you could validate that wait, there are some just to close the discussion that would be incredible if you could demonstrate that that is a very important question ok I can speak with you there are some people that say that traffic does not affect creep that is an assumption there is a very famous person like the almighty god of research I would not say the name but he affirms and gives some evidence that traffic load does not affect creep traffic does affect creep if you slow down the traffic maybe you magnify the impact of the creep exactly yeah