Skip to main content

Bridge Safety Monitoring

Final Report Summary - BRIDGEMON (Bridge Safety Monitoring)

Executive Summary:
In order to accurately assess the safety of a bridge it is important to have detailed knowledge of the loading which the structure is experiencing while in service. In addition to this, it is equally important to be able to accurately calculate the ability of the bridge to resist that loading. Once this information is available, advanced probabilistic methods can then be employed to calculate the ‘safety’ of the bridge.
It is often the case that existing bridges, which have been in service for a number of years, reach the end of their ‘design life’, i.e. the period of time for which they were designed to remain safely in service. Similarly, many bridges which are in service can become deteriorated, or can be subjected to unexpected loading situations which were not envisaged during their design. The safety of these bridges may be called into question. In such cases it is necessary to assess the ability of the structure to carry the loading that it is experiencing. Bridge owners/managers generally have fixed budgets available to them for carrying out repairs or maintenance. Many of the bridges in Europe and around the world are old and have already reached the end of their design lives. Obviously it is infeasible for road owners to replace all of these bridges, and therefore it is important that they are able to identify which bridges are safe to remain in service and which will require strengthening or maintenance.
For this reason it is necessary to carry out safety assessments for these bridges to try and evaluate the level of remaining capacity (often these bridges have been over-designed and have not been subjected to the conservative loading scenarios for which they were designed). Various methods of bridge assessment exist, depending on the level of information available on the actual loading that a bridge is experiencing or its ability to resist that loading. The more information that is available, the more accurate the calculation of the bridge safety will be. Very often there is not a great deal of information available and bridge safety is assessed using assumptions on the actual loading or resistance of the bridge. Where these assumptions are made they tend to be conservative in order to avoid over estimating the safety of a bridge.
As such, tools which allow the loading or the resistance of a bridge to be accurately measured are a great asset to road owners or infrastructure managers, allowing them to decide the most appropriate way to distribute their maintenance budgets. The BridgeMon project aimed to develop tools which can be used to allow more detailed information to be collected on the loading being experienced by a bridge in service along with its ability to resist that loading. Bridge Weigh-in-Motion (B-WIM) refers to the technology which uses measurements taken from a bridge as a truck drives over it, at full speed, to calculate the weight of that truck. Such a technology allows large amounts of information on the traffic loading on a particular bridge or on a particular section of a road network to be collected. BridgeMon aimed to improve the accuracy of the SiWIM B-WIM system, which is already commercially available for road bridges, and also to apply the B-WIM approach to railway bridges, allowing the weights of trains to be easily calculated. In addition, BridgeMon developed a ‘virtual monitoring’ approach which could be used to estimate the remaining fatigue life of steel bridges and hence provide accurate information on the resistance of the bridge to fatigue loading. Algorithms for identifying when a railway bridge has become damaged were also developed as part of the project.
The final results of the project, using in-field testing, showed that the accuracy of the SiWIM B-WIM system was significantly improved, with an accuracy class A(5) achieved for gross weights when the light vehicles, which were of little interest from a traffic load monitoring perspective, were excluded.
The virtual monitoring technique was successfully implemented and tested on a major steel bridge in the Netherlands, providing a powerful tool for the calculation of fatigue damage.
Finally, the Bridge-WIM technique was adapted to be suitable for railways and results showed that it could accurately calculate the weights of trains. A number of structural health monitoring algorithms were also developed for railway bridges, and while results could only be evaluated using numerical simulations, it was shown that these algorithms may be capable of detecting damage.
The results of the BridgeMon project have been very positive and all of the developments arising from the project will provide the SME partners with improved competitiveness in their relevant bridge monitoring industries.

Project Context and Objectives:
Traffic loading is one of the key parameters governing the design and assessment of bridge structures and road pavements. As such, the ability to monitor the actual traffic loading on a particular bridge or road can be of great benefit to infrastructure managers who, with limited budgets, are required to make informed decisions regarding repair or maintenance strategies for road infrastructure.
Bridges which are nearing the end of their design lives and have lost load-carrying capacity may still be safe to remain in service if it can be shown that they are carrying less than their original design traffic loading. In order to gather information on the traffic loading, it is necessary to weigh all of the vehicles traversing a bridge over an extended period of time. It is not feasible to stop every vehicle and statically weigh them – apart from being hugely inefficient, this results in biased statistical data as overloaded vehicle drivers will attempt to avoid being weighed. Therefore, the use of Weigh-in-Motion (WIM) technology, i.e. weighing vehicles which are travelling at speed, is the only efficient way to collect traffic loading information. This information can also be used for transport planning and pavement design and maintenance strategies among other things.
WIM technologies can generally be classed within one of two categories (i) strip or plate sensors embedded in a groove cut into the pavement or (ii) using sensors attached to an existing bridge. The latter is known as Bridge-WIM, and can be advantageous, as unlike traditional pavement-WIM systems it is portable. Furthermore, it does not require road closure during installation and maintenance because the sensors are located on the underside of a bridge, beneath the road. The use of a measured bridge response to calculate vehicle weights can also be indirectly beneficial as it provides information on the structural behaviour of the bridge. Changes in bridge response over time can provide an indication of damage in the structure.
The SiWIM® Bridge-WIM system is a commercially available system for the weighing of vehicles and is a product of the Cestel d.o.o. company in Slovenia. The system is used in many countries worldwide. The aim of Work Package 1 of the BridgeMon project was to employ a number of the latest theoretical developments in Bridge-WIM research into the SiWIM system in an effort to increase its accuracy and long-term stability. The accuracy of WIM systems can be classified in accordance with the COST 323 draft European WIM specification. The ambition of BridgeMon was to improve the accuracy of SiWIM by one accuracy class for any given bridge installation and to achieve Class A(5) accuracy, which would be extremely rare in the WIM industry and would therefore provide a significant advantage for the Cestel company.
The objectives of the BridgeMon proposal were to develop new bridge monitoring technologies:
Objective 1. To develop a method of using an existing bridge to weigh vehicles in motion with Class A accuracy;
Objective 2. To develop a novel method of monitoring steel road bridges, susceptible to fatigue damage;
Objective 3. To develop an accurate and cost-effective method of weighing trains in motion and
Objective 4. To develop an improved method of monitoring railway bridges.
Two types of bridges were considered in the proposal, road and rail. While they have much in common, these need to be considered separately as the markets are quite distinct and the nature of the traffic load is rather different (e.g. tracks are smoother than road pavements which reduces dynamics but the train’s weight is greater than a truck’s).
Objective 1 – Load on Road Bridges
The goal in the Road Bridge Load strand was to achieve a step change increase in the accuracy of ‘Bridge Weigh-in-Motion’ (Bridge WIM), the technology of weighing trucks while they are driving over existing bridges. According to the Comité Européen de Normalisation (CEN) draft standard on WIM accuracy of results is defined as one for which roughly 95% of gross vehicle weights can be found with certain accuracy. Objective 1 was to achieve, where possible, Class A(5), i.e. ±5% accuracy of gross weight for roughly 95% of all results, or at least considerable (at least one accuracy class) improvement compared to the pre-BridgeMon state-of-the-art.
There are many sensor technologies for weighing vehicles in motion available on the market – the so called Weigh-in-Motion (WIM) technologies – piezo ceramic, piezo-polymer, piezo-quartz, capacitive and bending plate. These sensors are embedded in a groove in the road pavement and can achieve accuracy Class B(10) – or B+(7) under ideal pavement conditions - according to the CEN draft standard on WIM. Cestel, the project leader, employs and develops the SiWIM® B-WIM system that uses an existing bridge, with sensors attached, as the weighing scales. The goal of the project was to improve the SiWIM® system so that it consistently achieves accuracy which is one accuracy class better, i.e. on appropriate bridges, to increase accuracy from class B+(7) to A(5), a level of accuracy that is ahead of all other WIM technologies.
Using an existing bridge as a weighing scale has many advantages over other WIM technologies. There is nothing on the road surface so there is no need for lane closures during installation or maintenance. There is also a major technical advantage in B-WIM in that the bridge is long enough to compensate for errors due to dynamic oscillation of the truck axles as they bounce and rock along the road. This bouncing and rocking motion of the truck is a problem for the other types of WIM sensor. As they are only a few cm wide, they cannot capture the exact weights, unless multiple sensors are installed at great cost. B-WIM systems are currently priced competitively compared to other technologies.
Objective 2 – Resistance of Road Bridges
The second objective of the BridgeMon proposal was to monitor and quantify the fatigue resistance of road bridges to traffic loading.
Practice in most European countries is to monitor bridges on a regular basis. Typically, a team of bridge inspectors arranges a visit every one to two years to every bridge on the road network. Regular inspections are mostly visual – the inspector looks for cracks or other signs of distress.
Increasingly and especially for large important bridges, there is a move towards Structural Health Monitoring, automated systems where critical elements in bridges of concern are monitored automatically.
Monitoring steel road bridges for the effects of fatigue is a particular problem. Fatigue damage often does not become visible until after many millions of cycles of stress over a certain threshold. Monitoring a bridge at points that are vulnerable to fatigue damage is a standard procedure. However, there are many other locations where it would be useful to monitor but bridges can be many hundreds of metres long and it is not practical to monitor everything on a continuous basis.
In BridgeMon, the concept of virtual monitoring would be developed. The Bridge WIM sensors would be used to calculate the weight of every truck that crosses. This would be used in turn to calculate the stresses at every point on the bridge, including points where there are no sensors by using an accurate computer model of the bridge. This would provide Bridge Engineers with stress histories for every point and would give them everything they need to calculate the remaining fatigue life for every element on the bridge, not just the elements where there are sensors. Virtual monitoring would provide a great deal of added value to a Structural Health Monitoring system, particularly for steel bridges where fatigue is often a primary concern. KALIBRA (Partner No. 2) and Corner Stone (Partner No. 6) have extensive experience of bridge monitoring and this innovation would give them a unique selling point to market that expertise internationally.
Objective 3 – Monitoring Load in Railway Bridges
Before BridgeMon, B-WIM was a road bridge concept and has not been successfully applied to railway bridges. While there are advantages when applying the Bridge WIM technology to railway bridges – the train is constrained to stay on the tracks for example – there are also major problems. Train dynamics can be large and the presence of sleepers and ballast between the track and the deck for many bridges makes it difficult to detect axles.
In BridgeMon, the Bridge WIM concept would be extended to railway bridges. The significant technical problems arising from train/bridge dynamic interaction would need to be overcome, most likely by solving the dynamic equations but allowing for the mass of the vehicle, a major scientific challenge.
The more practical problem of detecting the locations of wheels on the bridge will also be addressed. As trains are often electric, electrical resistance strain gauges on the track could be unsuitable for locating the wheels and fibre-optic sensors tend to be expensive. The potential solution would be piezo-electric sensors in the track.
Monitoring load with a bridge WIM system is new to railway bridges and would provide a new market opportunity for Adaptronica, already established in the industry of monitoring railway bridges. Current methods of monitoring train loads are cumbersome and slow. Furthermore, as the rail markets across Europe are deregulated, the track owner will have less control over the train operators. In this kind of environment, new methods need to be available for the track owner to monitor the compliance of the train operators with the legal weight limits. This would result in an increased interest in railway WIM in recent years.
Objective 4 – Monitoring Resistance in Railway Bridges
A new method of monitoring the resistance of railway bridges would be developed that exploits knowledge of train locomotive weights. Even with deregulation, there are only a finite number of locomotive models on the network and knowledge of the weight of these, combined with bridge measurement, can be used to identify a change in the bridge behaviour. A histogram can be plotted for the calculated weight of all locomotives of a given type which, given their consistent weight, will have a ‘sharp’ peak (small standard deviation). Any change in the peak of this histogram suggests a change in the bridge. This may be due to temperature, support conditions or stiffness. Either way, it provides valuable new data about the bridge condition which will be developed into a marketable product.
New algorithms would be developed of train/track/bridge dynamic interaction that would be subsequently used by Adaptronica for the optimal design of Structural Health Monitoring (SHM) systems – automatic sensor systems that monitor the bridge’s resistance continuously over time. Finally, a combined SHM and Bridge WIM system would offer the bridge owner better value for money – information on load and resistance from only one combined system. This would give Adaptronica its unique selling point to distinguish itself from competitors in the railway bridge monitoring market.
In summary, the project addressed the concept of ‘Research for the Benefit of SMEs’ by developing new scientific approaches that give our SME partners unique selling points in the Bridge Safety Monitoring market. Cestel would be given a Bridge WIM system that is more accurate and stable than any of its competitors. Kalibra and CornerStone would be given the Virtual Monitoring concept for the monitoring of steel bridges and Adaptronica would be given a Railway Bridge WIM system and improved methods of monitoring railway bridges.

Project Results:
WP1
The first objective of BridgeMon project was to achieve a step change increase in the accuracy of ‘Bridge Weigh-in-Motion’ (B-WIM), the technology of weighing trucks while they are driving over existing bridges. Thus, a number of features which have the potential to improve the accuracy and long-term stability of the SiWIM® system, were investigated. These improvements were grouped into the following categories:
• Improved axle-detection methods.
• Improved Bridge-WIM algorithms.
• Improved temperature compensation procedures.
• Improved data quality assurance.
Preliminary testing had been carried out in order to define appropriate procedures for the long-term in-field testing and validation of the new techniques. In order to assess the accuracy of the new methods a number of vehicles of known weight were required for testing. These vehicles were statically weighed with assistance of police, and their axle spacings were recorded to provide a baseline against which B-WIM system accuracy can be assessed. In order to assess the accuracy of temperature calibration procedures, the temperature of the bridge was measured on at least two points within the deck and correlated to the passage of each of the test vehicles. In addition, two of the proposed Bridge-WIM algorithms required additional instrumentation on the bridge, compared to the standard configuration. Measurements at the quarter and three-quarter points of the bridge span were performed, in addition to the standard procedure of measuring strains at the mid-span.
It was agreed within the BridgeMon consortium that testing of these methods in-field should be carried out in two distinct phases (i) initial testing using a sample of statically weighed trucks whose static weights are given to the research providers and (ii) blind testing using statically weighed trucks whose static weights are unknown to the research providers. The accuracy of all of these methods were compared to the accuracy of the current procedures used within the SiWIM® system and were implemented into the SiWIM® software accordingly.
Improved axle detection
The accuracy of the weights calculated by a B-WIM system is highly dependent on the ability to accurately identify the number of axles on a vehicle and the spacings between them. The SiWIM® system identifies axles from peaks in the measured signals. This is often challenging for certain bridge types, particularly beam-slab bridges.
Firstly, an alternative instrumentation strategy for axle detection on concrete beam-and-slab bridges was developed. A bridge of this type near Vransko in Slovenia (VA0468) was selected for numerical modelling and measurements of the improved axle detection algorithms. The bridge is a simply supported, pre-stressed concrete beam and slab bridge, which is a typical form of construction for highway bridges. B-WIM installations in the past have shown problems in identifying axles from measured signals for this type of bridge. A detailed 3-D numerical model of the bridge was developed to identify measurement locations most sensitive to the passage of axles, allowing the optimal sensor locations for reliable axle detection. The results of the modelling were successfully verified on the VA0468 bridge. In succession, instrumentation of the VA0030 beam-and-slab bridge, which was used for long-term testing, was modified accordingly.
Modified instrumentation coupled with improved techniques for processing of axle detection signals, along with some other modifications to the current axle detection algorithms, resulted in extremely accurate axle detection. Only 3 of 542 axles were miss-detected, in one of the 122 reference vehicles on the VA0030 beam and slab bridge, with only 3 out of 778 axles, in 3 of 178 vehicles on the VA0028 culvert being miss-detected. Furthermore, detailed inspection of measured records suggests that most miss-detected axles were not due to poor axle detection but because the axles were extremely light or even lifted (they were lowered only during the static weighing).
Improved Bridge-WIM algorithms, including temperature and velocity compensation
A vast number of other enhancements were implemented within the system with updated algorithms for calculating the bridge influence line, calculating vehicle weights, temperature/velocity calibration and improved approaches for quality assurance of measurements.
Testing of the various concepts and approaches was carried out using in-field measurements on three different bridges in Slovenia. In addition to the bridge VA0468, which was used to identify improved axle detection instrumentation strategies, two other bridges were instrumented: one a culvert spanning 6m (VA0028), and one skewed beam and slab bridge spanning approximately 20m (VA0030). A number of different phases of in-field testing were carried out during the course of the project. The static weights of only a limited number of the measured vehicles were disclosed to the research providers.
An already robust non-linear algorithm for calculating the influence line of a bridge directly from measurements has been modified to allow an accurate representation of the bridge influence line to be obtained with minimal calibration requirements.
A number of the most advanced Bridge-WIM algorithms arising from recent research in this field were tested. While some of the algorithms were shown to improve accuracy, many were found to be difficult to implement and were less practical than the traditional approach. During the course of the project it was discovered that traditional Bridge-WIM approach of adding the contribution of each of the sensors located across the width of the deck is inherently incorrect and will reduce the potential accuracy of the calculated truck weights. It was found that the influence lines for each position across the width of the bridge vary in length and shape and as such, combining the response of all sensors is not the best approach. In addition to this, it was found that due to the fixing of a given sensor to the bridge along with any localised defects or cracking in the vicinity of the sensor, the relative magnitudes of the measured responses did not reflect the expected transverse distribution of strain. Hence, calculations were biased towards certain sensors which were always giving the largest readings, irrespective of the transverse position of the truck. An automated procedure was developed to correct the measured values in each sensor so that relative magnitudes represented the expected transverse distribution of strain. It was shown that the most accurate and reliable weight predictions were obtained when only the sensors underneath the lane of traffic were used for weighing. This represents a change from the traditional Bridge-WIM instrumentation approach; however it was shown to increase accuracy significantly.
Auto-calibration routines were developed to account for the effects of temperature and vehicle velocity. The effect of temperature was seen to be less significant than was expected at the outset of the project. Full numerical modelling of dynamic vehicle-bridge interaction was used to demonstrate the effects that vehicle velocity can have on the bridge response measured by a Bridge-WIM system for common truck types. The effects of velocity were shown to be particularly important on shorter bridges and a velocity auto calibration procedure was developed which mitigates the effects of velocity for different vehicles.
Improved data quality assurance
In addition to all of the developments towards improving the accuracy of calculated weights, additional features have been developed to flag potential errors in the data. Various measures are used for these quality assurance procedures. The most basic approach consists of simple rules which, based on unusual characteristics within a particular record, assign a flag to a vehicle. A more advanced ‘kernel density’ approach is used to identify axle configurations or weight configurations which are very rare and are likely to be errors. In addition, a quality estimate (QE) can be obtained by comparing the measured signal to the reconstructed signal from the Bridge-WIM algorithm. The transverse position of the vehicle can also be used to indicate how accurate a weight prediction is likely to be. Statistical monitoring of data from a particular Bridge-WIM site over time can be used to identify any loss of calibration or problems with the system. Most notably, real time monitoring of the measured values in each of the sensors has shown that for long term measurements, sensor drift can be an issue and, if it appears, must be corrected. Including this drift correction has resulted in more robust measurements over time and enables SiWIM® to be used to carry out accurate long term monitoring of traffic loading.
Results
The final section of the report D1.3 compares the accuracy of the calculated weights from the SiWIM® system after the inclusion of the developments arising from BridgeMon to those obtained using the algorithms available within SiWIM® prior to the commencement of the project. Results from two test bridges are presented and the accuracy of the system is assessed at different stages of development, ranging from the pre-BridgeMon version of the system, to that with all of the latest developments included.
The first of the two bridges is the 6m culvert (VA0028) on which 178 vehicles had been statically weighed. The long-term accuracy classification of the pre-BridgeMon version of SiWIM® on this bridge was E(35) which was poor (it should be noted that prior to BridgeMon SiWIM® was used primarily for 1-week to 1-month measurements). After the inclusion of the BridgeMon developments the accuracy class increased by four accuracy classes to C(15). This represents a significant improvement on the original results. A great deal of work was carried out to try and improve the accuracy class beyond C(15), however, no further improvement could be made. Reference is made to the unevenness of the pavement, and in particular a bump which occurs directly before the entry to the bridge. Vehicle dynamics induced by an uneven pavement can lead to axle forces, which are different to the static values, being exerted on the bridge. This is more of a problem on shorter bridges where the vehicle is only on the bridge for a short period of time and the overall vibration of the vehicle does not have a chance to cancel out like it would on longer bridges. Reference is made to a previous study which showed, in similar circumstances, that resurfacing the pavement resulted in much more repeatable strain readings and improved the accuracy of Bridge-WIM calculations significantly. The inability of SiWIM® to achieve an accuracy class beyond C(15) on the 6m culvert appears to result directly from the uneven pavement at the site and serves to demonstrate the importance of, not only choosing an appropriate bridge, but also ensuring that the road profile is sufficiently smooth in order to achieve a Bridge-WIM installation of the highest accuracy. A site with much smoother pavement has already been instrumented to test the possibilities of the new SiWIM® system in more ideal road conditions.
Results from the VA0030 beam-and-slab bridge, on which 122 vehicles were statically weighed, are also presented in this report. For the correctly identified vehicles the pre-BridgeMon version of SiWIM® attained an accuracy class of D+(20) on this bridge. After the implementation of the latest developments SiWIM® improved by two accuracy classes to class B(10). It is also noted that when the lightest trucks, which were of less interest from a traffic monitoring perspective, were removed from the accuracy classification, the system achieved class A(5) accuracy for gross weights. The ability to achieve A(5) accuracy for gross weights is extremely rare in WIM market and represents a huge advancement in the field of Bridge-WIM, particularly considering that the results were achieved over such a long term monitoring period.
The work carried out in WP1 of the BridgeMon project has lead to considerable improvements in the accuracy of the SiWIM® system and will certainly improve the ability of Cestel to compete with other WIM technologies in the field of traffic load monitoring.

WP2
General Overview
When trucks drive across bridges, the structural members carrying the weight of the vehicles become stressed. When a truck leaves the bridge these members return to their original stress level. This continuous cyclical stressing and unstressing can be problematic, particularly in the case of steel bridges. When steel is subjected to this repetitive loading, very small cracks can develop and over time can lead to larger structural cracks which can compromise the safety of the bridge. This type of damage is known as fatigue damage and is very often one of the most critical aspects to be considered when examining the safety of steel bridges.
BridgeMon has developed a novel technique for monitoring fatigue damage in steel bridges. This newly developed ‘virtual monitoring’ approach uses a Bridge-WIM system which is installed on a bridge to monitor the actual traffic loading at the site. The ‘virtual monitoring’ technique also takes advantage of the fact that a Bridge-WIM system measures strains on the bridge and hence provides some information on how the bridge behaves when subjected to traffic loading.
The general idea behind the ‘virtual monitoring’ approach is as follows:
1. Install a Bridge-WIM system on the bridge of interest.
a. Sensors should be installed in the usual manner so that the Bridge-WIM system can accurately calculate the weights of trucks at the site.
b. Additional sensors can be installed as necessary at some locations of interest.
2. Develop a computer model of the bridge.
a. Using the information on the truck weights and the structural behaviour of the bridge, recorded by the Bridge-WIM system, the computer model should be calibrated so that it accurately represents the behaviour of the bridge.
3. Monitor the traffic loading at the site.
4. Use the observed traffic at the site to simulate traffic flow.
a. The observed traffic can be used to calculate future traffic flow trends.
b. Traffic can be simulated for various periods into the future, e.g. 10, 20, 50 years.
5. Use the simulated traffic flow to evaluate the structural response of the bridge.
a. The computer model of the bridge should be subjected to the simulated traffic flow to calculate the range of stresses induced in the bridge over time.
b. The structural response at any location on the bridge where the response is accurately described by the computer model can be evaluated.
6. Evaluate the fatigue damage at critical locations.




Using the approach outlined above it is possible to estimate the remaining fatigue life of a bridge as the computer model (assuming it is sufficiently accurate) can be used to evaluate the level of fatigue damage at any location, i.e. not only at the locations which have been directly monitored on-site. This is an extremely useful tool for bridge owners who are dealing with steel bridges where fatigue is of primary concern. It avoids the necessity for the installation of vast amounts of sensors at various locations of interest and also provides the user with detailed information on the traffic at the site. While it is noted that the virtual monitoring approach developed in BridgeMon only addresses fatigue and does not account for other mechanisms which may cause problems in a bridge, it is often the case in steel bridges that fatigue is of primary concern and the virtual monitoring technique is a valuable tool in these instances.

Implementation on Steel Bridge in the Netherlands
During the course of the BridgeMon project the virtual monitoring concept was tested on a bridge in the Netherlands. At the outset of the project, discussions were held with Rijkswaterstaat (the Dutch authority responsible for road infrastructure) to decide on the bridge that could be utilised for testing as part of BridgeMon. Rijkswaterstaat suggested a bridge that was of particular interest to them and allowed the BridgeMon consortium to use this bridge to implement the virtual monitoring technique. Close contact was maintained with Rijkswaterstaat for the duration of the project.
At the outset, it was expected that the virtual monitoring technique would be developed for typical steel bridge types which could be represented by relatively straightforward computer models. It was envisaged that the properties of these models could be automatically calibrated using the measured data from the Bridge-WIM system. The bridge which Rijkswaterstaat provided for testing of the virtual monitoring system was a cable-stayed bridge, whose structural behaviour was complex and unique. The development and calibration of the model for this bridge was far more intricate and complicated than had been planned at the outset. In the interest of maintaining good relations with Rijkswaterstaat, and potentially gaining a future client, the virtual monitoring approach was modified slightly in order to allow it to work for such a complex structure.
Due to the complexity of the structure being considered, and the fact that it was unsuitable for accurate Bridge-WIM calculations, an adjacent simply supported span was used for the installation of the Bridge-WIM system. This ensured that accuracy of the calculated traffic loading would not be compromised. In order to monitor the cable stayed part of the structure it was then necessary to install additional sensors on this part of the bridge so that the computer model could be validated (it was initially expected that the Bridge-WIM sensors could also serve this purpose, however this was not practical for a bridge of this nature). Due to the complexity of the cable-stayed bridge being examined, an auto-calibration procedure whereby the computer model would automatically be calibrated to match the measurements was not practical. Instead, for a bridge of this type, it was necessary to manually calibrate the model to ensure that all of the material and geometrical properties used in the computer model were appropriate. It was initially intended that a reasonably straightforward template computer model could be automatically calibrated from the measurements, however when dealing with cable-stayed bridges the specific structural properties of a given bridge can vary dramatically and as such it was not feasible to develop a method of automatically calibrating the computer model.

Initially, preliminary computer models were developed to assist in identifying the most suitable locations for the installation of sensors. Strain gauges were installed on the bridge in October 2013, with the Bridge-WIM system recording traffic information for 38 days between the end of October and the beginning of December. The information that was recorded was then used to calibrate the detailed computer model of the cable-stayed bridge. It is worth noting that due to the fact that two separate data acquisition systems were installed (i.e. the Bridge-WIM system on the simply supported span and another system for the strain gauges on the cable-stayed span) some difficulties were encountered when trying to identify the equivalent strain records corresponding to the passage of a vehicle(s) for the two systems.
Rigorous cleaning of the data was carried out in order to remove any errors in either the recorded traffic loading or strain signals. The cleaned traffic data was then examined and detailed analysis of the data was carried out to identify trends in the traffic flow which could be described using probabilistic methods. This allowed 15 years worth of traffic loading scenarios to be generated based on the nature of the traffic observed on the bridge during the 38 day monitoring period.
Using the 15 years of simulated traffic, long run simulations were carried out whereby the traffic was run across the computer model of the bridge and the stresses induced in every member on the bridge were recorded for the 15 year duration. An algorithm was developed which was capable of analysing the fatigue damage induced in particular parts of the bridge by counting the number of stress cycles in particular elements (i.e. the amount of loading/unloading experienced by various structural members on the bridge). The level of fatigue loading calculated for these structural members was compared to the relevant fatigue resistance for the particular member (the design codes specify the number of stress cycles which particular structural members can endure before they suffer fatigue failure).
The fatigue damage caused by the traffic was then compared to that induced by the loading which the Eurocodes specify when designing steel bridges for the effects of fatigue. It was shown that the fatigue damage induced by the code loading was far in excess of that induced by the simulated traffic.

Conclusion
While the final format of the virtual monitoring approach developed within BridgeMon was slightly different to that which was initially envisaged the implementation on the Dutch bridge showed great promise. The bridge owners were very interested in the outcome of the project and were closely involved at all stages.
The final 'virtual monitoring' package which has been developed as part of the project essentially consists of two separate parts:
(i) Bridge-WIM system
(ii) Virtual monitoring service
The Bridge-WIM system monitors the traffic loading on the bridge while simultaneously measuring the structural response of the bridge to this traffic loading. Where a fatigue damage calculation is required the 'virtual monitoring' service can also be availed of, whereby the computer model of the bridge is developed and calibrated using the measured data. The algorithms which have been developed to simulate traffic based on that observed during the monitoring period can then be applied and a fatigue damage calculation can be carried out for any location of interest on the bridge.
It is expected that the ability to carry out this virtual monitoring will provide bridge owners with an extremely useful tool for monitoring fatigue levels on steel bridges and will allow them to make better informed decisions towards the allocation of their maintenance budgets.





WP3
Structural Health Monitoring of Railway Bridges
As part of BridgeMon a number of different structural health monitoring (SHM) algorithms were developed and tested for railway bridges. The overall aim was to develop algorithms which would be capable of indicating when damage has occurred to a bridge. These SHM algorithms would take advantage of the fact that the railway Bridge-WIM system measures the structural response of the bridge while it is in operation. A number of alternative approaches were tested in order to identify a suitable method of damage detection in railway bridges, with the Nieporęt Bridge in Poland being used for testing.
A detailed computer model was developed for Adaptronica which was capable of modelling structural response of the bridge, while allowing for the fact that the train itself vibrates as it is crossing the bridge. Through complex interaction with the railway track, any stone ballast under the railway line, the sleepers, and the bridge itself, the overall structural response of the bridge can be modelled. Software was developed which allows the properties of the computer model to be adjusted to represent any similar truss bridge for which a detailed understanding of the response may be required.

Most SHM approaches for bridges examine the vibrating response of the structure to identify the natural frequency at which it vibrates. Generally, when the bridge becomes damaged there is a change in the way that the structure vibrates and this often manifests itself in a change in the natural frequency. The majority of the state-of-the-art SHM approaches for bridges examine the vibrations of the bridge over time to try and identify if there is any change in the natural frequency, which may be caused by damage.
The detailed numerical model served as the basis for the testing of various SHM approaches which were developed during the BridgeMon project. The reason for using this model was that it could accurately represent the complex interaction which causes the bridge to vibrate as a train passes over it. Using the model, damage to the bridge could artificially be simulated and SHM algorithms tested to see if this damage could be identified by examining the response of the damaged bridge. The model also serves the purpose of allowing informed decisions to be made as to the optimum instrumentation strategies for an SHM system, simulations can be carried out to show where on the bridge the measurements are most likely to change if it becomes damaged. While in-field measurements were taken from the bridge during testing, these could not be used to try and identify damage to the bridge (it was not possible to damage an in-service bridge). These measurements served the purpose of ensuring that the actual measurements taken on the bridge were consistent and therefore that the SHM algorithms would not falsely identify the bridge as having become damaged due to any measurement noise that may be picked up when monitoring the bridge.

Using the computer model, three approaches to try and detect damage induced in the bridge were tested. The first two were conventional approaches with the third approach being an alternative on the traditional approaches which was developed as part of BridgeMon:

1) Identification of a change of the measured strain response of the bridge due to damage
2) Identification of a change in the natural frequency of the bridge due to damage
3) Change in the frequency content due to damage using wavelet analysis

The first method mentioned above takes advantage of the fact that the locomotives at the front of trains are generally standard configurations and weight and only a few variations are found on a given section of railway. As such, the strains induced in the bridge should be very similar when each time one of these standard locomotives is on the bridge. It is known that for certain types of structure (including the Nieporęt Bridge) that the strains induced in one section of the bridge are independent of damage at a different location, so unless the measurements are being taken at the location that becomes damaged it is unlikely that the damage will be identifiable. A number of simulations were carried out, whereby different levels of damage were modelled in different parts of the bridge and the results, as expected, confirmed that damage could be detected using the measured strains from the bridge, but only when the damage occurred at a location from which the strain response was being measured. Due to the fact that the railway Bridge-WIM system measures strains on the bridge this approach could still be useful in identifying if the structure becomes damaged at any of these measuring locations.

The second approach which was examined aimed to identify any change in the natural frequency of the bridge when it became damaged. In this case, the acceleration of certain parts of the bridge was measured during the simulations instead of strain. A Fourier transform was then applied to the recorded acceleration measurements from the model and any change in the natural frequency when the bridge became damaged was noted. It was shown that while a change in the natural frequency of the bridge was identifiable after damage, the change was reasonably small (about 5%) even in the cases where the bridge had become significantly damaged.

The final approach which was examined was a slightly more sophisticated method. Again, acceleration measurements were required for this method and a 'wavelet transform' was used to try and identify any small changes in the measured response of the bridge after it was damaged. In basic terms, the wavelet transform examines, in detail, the shape of each part of the measured bridge response and then this can be used to identify if/where it changes due to damage. The proposed algorithm was shown to be capable of identifying damage to the bridge, with the particular advantage of this method being that the actual location of the damage could also be roughly estimated. This approach was shown to be superior to the previous two approaches.

In summary, a detailed numerical model which allows simulations to be carried out to identify the optimum instrumentation strategy for SHM purposes has been developed. A number of SHM algorithms have also been developed and tested. These algorithms have been shown to be capable of identifying when a bridge has become damaged. SHM technologies are still in their infancy and without the ability to take measurements from a bridge before and after damage has occurred it is not possible to truly field-test these algorithms. As such, detailed computer modelling is the most appropriate way to test various methods for identifying damage to bridges.

Potential Impact:
Promoting the use and dissemination of the project results is a key objective of the Seventh Framework Programme (FP7) and the document “Plan for the Use and Dissemination of the Foreground - PUDF” is an obligatory document in the “Benefit for SMEs Progamme” and as such was created in the BridgeMon project as well as Deliverable D6.3.
During the project he consortium determined how the results ("Foregroud) of the BridgeMon technologies will be disseminated during the project, as well as how they will be exploited following project completion. The consortium:
• Identified and defined exploitable results
• Identified and defined the need and the direction for further post-project development
• Developed a plan for the exploitation of results
• Disseminated the results during the project
• Developed a plan for further dissemination in the post-project phase

In addition, the consortium explored the potential socio-economic impact of the results, the target groups for dissemination and exploitation activities, as well as the contributions to industry standards or policy developments that the BridgeMon project could possibly make. Moreover, no potential risk stemming from the results which could impinge negatively on the project has been identified.
The information and results are based on:
• the feedback from all consortium partners
• market research conducted to identify potential outlets for the Foreground and an initial business plan
• a summary of performed dissemination activities. A list is provided and discussed.

Results/knowledge that can be disseminated
By default, any result of the project is to be treated as confidential and can be disseminated only after overall agreement of the project partners. This approach allows partners to have the full control and make sure that no “exploitable” information is disclosed before being protected. Such an approach should also allow the basis for the technical data that RTDs can use to prepare scientific communication to be defined (which will be revised and approved by the consortium in any case).
The protection of the intellectual property rights (IPR) is of invaluable importance for the participating partners. Therefore, the non-confidential results and information on the project eligible for public dissemination needs to be identified. The following is a non-exhaustive list of non-confidential information and documentation that is being used for project dissemination purposes to date:
• project newsletters (3 issued)
• The specific role played by each partner; and the concepts underpinning the project and its goals of the project are described.
• Information on the expectation of the partners in the project
• Unique Selling Points
• Opportunities and market potential of the technology
• Socio-economic impact
• Scientific publications, posters and presentations previously approved by the consortium
• All the information approved by the consortium to be published on the project website

In each publication, as well as in the all press releases and other dissemination actions, the BridgeMon project is briefly described and the EC grant is duly noted in the acknowledgement with the following sentence being quoted: “The research leading to these results has received funding from the European Community’s Seventh Framework Programme [FP7/2007-2013] under grant agreement nº 315629”.
Different categories of audience need to be targeted to ensure that the developed technologies have a far-reaching impact:
• Professionals / Industrial players covering the whole supply and value chains:
o SMEs, large companies and Industry Associations in road/bridge construction
o Bridge inspection/monitoring companies
• Public entities:
o Road infrastructure owners/managers
o National roads administrations
o Railway infrastucture owners/managers
o Town planning authorities
o Police/truck weight enforcement bodies
o Local governments
• Scientific community
• General public / road users

Communication Strategy
The message and communication channel has to be adapted to each previously mentioned target audience and in particular the level of technical and scientific detail needs to be adjusted for each target.
For the general public audience, the key message focused on the socio-economic impacts of BridgeMon project, paying particular attention to communicate the sociological benefits of developing technologies which allow road and railway traffic to be closely monitored along with the techniques developed to monitor the health of bridges, all of which results in a safer, more sustainable, transport system.
In terms of the communication strategy for effectively reaching and stimulating the industry, an entirely different strategy with different goals was pursued. To this end, the Unique Selling Points (USPs) of the technology form the core message of industry communications, which also present the basic concepts of the technologies and emphasize their novelty. The key is to stimulate interest in the technology, thus promoting its uptake. Moreover, it was important to provide a solid technical and scientific proof of concept in order to substantiate the claims that are made.
The dissemination channels for approaching industry are quite diverse. Face-to-face communications at trade fairs and press releases in technical journals and on the Internet have been mostly used. Wider dissemination activities included workshops organized in each of three participating countries at the end of the project.
The messages, communication channels, as well as the role of each consortium partner in the dissemination activities in keeping with their position in the supply and value chain is address later on.
Role of partners in the dissemination
Each of the SMEs and RTDs of the BridgeMon consortium plays different and synergistic core roles in the project and in the dissemination and use of the results, which are carefully balanced to maximize the impact of all activities. Indeed each partner has different core competences, different sectors of application, different geographic locations and they were selected as such in order to maximize the impact of the BridgeMon project. Here we only report the general strategy for each type of organization, the details of the disseminations performed per each partner are presented in the final report and the specific use expected by each partner dictates the way and what way the audience of the disseminated information receives it. The role of each partner is also connected to each target group and relevant channel.
SMEs were in charge of disseminating the results among their current network of customers, websites, brochures, in the key events and trade fairs of their sector.
RTDs were expected to disseminate scientific and technical information (after agreement from the SMEs). Even though general information can be used in most articles, when it comes to the technical component, each RTD partner is responsible for publications in its own area of expertise and in this way the RTD partner keeps the benefit of its own work.
In broad terms, the following segmenting has been made in terms of dissemination effort among the partners:
• Cestel focused on contacting relevant stakeholders in Slovenia and internationally.
• Kalibra and Cornerstone agreed to contact the relevant Dutch stakeholders and Industry professionals.
• Adaptronica were responsible for contacting the relevant Polish stakeholders in the railway industry.
• ZAG and ROD were primarily responsible for the development of scientific and technical publications (with prior agreement from the SME partners).
• ZAG (with input from all partners) were responsible for the publication of the content on the project website and the project newsletters.
• All partners attended relevant trade fairs/conferences and presented information on the results of the BridgeMon project.

Report on Dissemination Activities carried to date
The BridgeMon consortium has performed a total of 17 dissemination activities in addition to the non public face to face meetings with customers, which are not discussed individually as these are confidential information), the details of each can be found in the final report under 3 broad categories:
• Scientific publications (peer reviewed) - one is currently under review
• Paper in Proceedings of a Conference/Workshop
• Dissemination activities:
o Oral presentation to a scientific event
o Posters
o Organization of workshops (3 events)
o Publications

The dissemination tools developed in order to support the dissemination effort, as well as the different dissemination channels used are described below.

1. Project logo
A project logo was designed in order to assist in branding the project and which could be featured on all project communications and documents.

2. Project website
The public website was designed, built and made available on-line in month 1 of the project. It has the official address bridgemon.fehrl.org. Due to development of completely new web platform the updated web site is physically located on the address bridgemon.zag.si. When the official address bridgemon.fehrl.org is selected the user is automatically redirected to the new site. The project website has also been extensively disseminated by all partners. The website presents general information and updates about the project, its objectives, project partners, referencing FP7 and Research for the benefit of SMEs programme, and provides interested parties with a means for getting in touch with the consortium. It also gives an up-to-date access to the project’s progress and press releases. All the project partners have linked the BridgeMon website from their company’s websites to increase the visibility of the project, and reciprocally the reader can find the link to the partners’ websites in the project public website.

3. Project Leaflet, newsletters and posters
Three newsletters were issued during the project, in October 2013, in June 2014 and in November 2014. These newsletters were compiled by RTDs ZAG and RODIS and published on the projects web site: http://bridgemon.zag.si/
The first issue provided an overview of the project and focused on the progress that had been made at the end of the first period. The second issue contained an invitation to the final workshops of the BridgeMon project and was aimed at relevant stakeholders. This issue also provided an update on some of the work that had been carried out during the project. The final newsletter was published at the end of the project and summarises and widely disseminates the main results of the project.

4. Mail Shots
A number of different mail shot packages are being sent to users. The product being discussed in each mail shot depends on the potential client addressed. Users who were targeted by such mail shots are: Bridge/Pavement owners, Truck overloading law enforcement, Road and Railway bridge owners/managers. Mail shots contained Cover Letter and Newsletter(s).

5. Face-to-face meetings
In terms of the on-site visits that were conducted over the course of the project, the results of these meetings are confidential and therefore no company names are mentioned. Each SME partner had at least half a dozen of face-to-face meetings with their potential customers.

6. Conferences, exhibitions and tradeshows
The project was presented in key conferences:
• International Association for Bridge and Structural Engineering (IABSE 2013)
• XIV International Winter Road Congress
• Intertraffic Amsterdam 2014
• Transport Research Arena (TRA 2014)
• North American Travel Monitoring Exposition and Conference (NATMEC 2014)
• Civil Engineering Research Association of Ireland (CERI 2014)
• 26th Australian Roads Research Board Conference (ARRB 2014)
• 9th Austroads Bridge Conference (ABC 2014)
• 8th Croatian Conference on Road Maintenance (Održavanje cesta 2014)
• National Seminar on the Renovation of Steel Bridges
• 6th International Conference on Mechanics and Materials Design
• 12th Slovene congress on roads and traffic
Two papers in proceedings of a conference/workshop have been published:
• Multiple Equation Bridge Weigh-in-Motion” written by Eugene OBrien, Peter Favai, Robert Corbally (RODIS and ZAG) and published in the IABSE Symposium Report, IABSE Symposium, Rotterdam 2013: Assessment, Upgrading and Refurbishment of Infrastructures (available at: http://www.ingentaconnect.com/content/iabse/report/2013/00000099/00000012/art00006 )
• BridgeMon: Improved Monitoring Techniques for Bridges” written by Peter Favai (CESTEL), Eugene Obrien (RODIS), Aleš Žnidarič (ZAG), Hans van Loo (CORNER STONE), Przemyslaw Kolakowski (ADAPTRONICA), Robert Corbally (RODIS) and published in Proceedings of Civil Engineering Research Association of Ireland Conference, Queen's University Belfast, Ireland, (CERI 2014).

7. Scientific publications
One scientific publication has been made so far: »Determination of the optimal correction factors in the response analysis of the bridge weigh-in-motion system” written by Aleš Žnidarič, Goran Turk, Eva Zupan, all from ZAG, is currently under review in the journal Engineering Structures.
More publications are in progress.

8. LinkedIn
A BridgeMon web profile has been set up on the LinkedIn networking site which all partners have joined. The page can be accessed at https://www.linkedin.com/groups/BridgeMon-5033238. This group page can be followed by any interested party and will generally be suggested to them by the website.

9. Workshops
Towards the conclusion of the project three demonstration events were organised to disseminate the results of BridgeMon to stakeholders and relevant parties. These demonstrations were held in Poland, the Netherlands and Slovenia. These were arguably the most important events for showcasing the outcomes of the project to the most likely future clients. More details on the specifics of each of these demonstrations can be found in the demonstrations summary report (deliverable D4.2) and in PUDF (deliverable 6.3)

10. Other dissemination
Furthermore, identification and contact of Project Coordinators of relevant and complementary EC and regional research initiatives and other initiatives was done in order to explore ways to interact, build upon and complement each other’s efforts (such as TRIMM). In addition, other public or private organizations were identified and contacted in order to explore avenues for cooperation, depending on their capacity to enrich the results and impact of BridgeMon, and maybe participate in the post-project research efforts to enlarge the field of application of the new technologies). The IPR help-desk was also contacted in order to get guidance on the routes that were most suitable for the protection of the foreground.

Planned Post-Project Dissemination Activities
Important upcoming events are tracked carefully by the consortium and discussed via project meetings and e-mail/phone communication. The participation of a consortium member at any event is announced in advance on the website to allow anyone who would like to meet with the BridgeMon partners, or learn more about the project, to know where to go. Regarding future disseminations that are planned after the project end, some partners will attend two international conferences (abstracts for paper were already submitted):
• 12th Slovene congress on roads and traffic, Portorož; 22/04/2015 and 23/04/2015 (ZAG and CESTEL will attend it)
• 6th International Conference on Mechanics and Materials Design, Azores, Portugal; 26/07/2015 to 30/07/2015 (RODIS will attend it)
To date, one scientific publication authored by ZAG and RODIS regarding the main public results of the BridgeMon project is under review, but more are planned in the future.

Dissemination activities will be essential even beyond the end of the project to ensure a successful uptake of the technologies on the market. Such activities as well as any additional marketing done by the SME partners to promote the developed technologies will be progressive from now on. It is expected that the frequency of dissemination activities will increase as the industrial partners will be able to present their products (Class A Bridge-WIM system, 'virtual monitoring' technology for bridges and railway Bridge-WIM and structural health monitoring system) to the market. The demonstration events for potential clients were carried out towards the end of the project, while wider marketing activities are among post-project activities.

Plan for the Exploitation of the Results
In the BridgeMon project, “Use” refers to the direct or indirect utilisation of results in further research activities other than those covered by the project, or for developing, creating and marketing the developed traffic load monitoring and bridge health monitoring algorithms of BridgeMon, or for creating and providing a related service. The results can be used directly by the partners in the project or indirectly through licensing to third party. FP7 participants are required to use the generated results, which must be properly addressed and planned. Foreground that can be applied industrially or commercially must be protected adequately and effectively. Only non-confidential data relating to the use of the foreground are presented in this Section of the Final Report.


Description of the Exploitable Results of the Project

As a result of the BridgeMon project a number of developed algorithms and technologies have been identified as having exploitable market potential for the SME partners.
1. Class A Bridge-WIM System
This commercially available Bridge-WIM system has been significantly improved during the course of the BridgeMon project. Improvements have taken place in a number of the algorithms used within the system to calculate the weights of vehicles passing over an instrumented bridge. The algorithms used to identify the axles on vehicles, along with the distances between them have been noticeably improved along with updated algorithms for calculating the bridge influence line, calculating vehicle weights, temperature/velocity calibration and improved approaches for quality assurance of measurements.
Testing the system using in-field measurements taken over a period of one year has shown that the accuracy of the system has improved by two accuracy classes on one bridge and four accuracy classes on another, with the most important result lying in the fact that the system was shown to be capable of obtaining class A(5) accuracy for gross weights when the heavier vehicles are considered (Class A(5) is the highest achievable accuracy classification for a WIM system).
2. ‘Virtual Monitoring’ Approach
The novel concept of ‘virtual monitoring’ has been developed and tested during the BridgeMon project. This approach combines a Bridge-WIM system, which monitors the traffic loading, with a detailed computer model of a bridge, which can accurately represent the structural behaviour of the bridge. Combining the two allows an estimate to be made of the remaining fatigue life of the bridge using detailed information recorded on site. This is an extremely useful tool for bridge owners who are dealing with steel bridges where fatigue is of primary concern. It avoids the necessity for the installation of vast amounts of sensors at various locations of interest and also provides the user with detailed information on the traffic at the site.
The final 'virtual monitoring' package which has been developed as part of the project essentially consists of two separate parts:
(i) Bridge-WIM system
(ii) Virtual monitoring service
The Bridge-WIM system monitors the traffic loading on the bridge while simultaneously measuring the structural response of the bridge to this traffic loading. Where a fatigue damage calculation is required the 'virtual monitoring' service can also be availed of, whereby the computer model of the bridge is developed and calibrated using the measured data. Algorithms which have been developed to simulate traffic based on that observed during the monitoring period can then be applied and a fatigue damage calculation can be carried out for any location of interest on the bridge.
It is expected that the ability to carry out this virtual monitoring will provide bridge owners with an extremely useful tool for monitoring fatigue levels on steel bridges and will allow them to make better informed decisions towards the allocation of their maintenance budgets.
3. Railway Bridge-WIM System and Structural Health Monitoring Algorithms
During the course of the BridgeMon project the first commercially available Bridge-WIM system for calculating the weights of trains in motion has been developed. This system is based on a similar concept to that used to weigh trucks on road bridges; however the algorithms have been updated where necessary to adapt the system to railway applications. As the track owners in some countries are being separated from the train operators, the owners need to know the weights of the trains so that they are not overweight and as such, the development of such a system should offer a product with a unique selling point in the railway market. In addition to the system for monitoring train weights, a number of structural health monitoring algorithms have been developed. With further testing and development of these procedures for identifying damage to railway bridges it is expected that the combined traffic load and health monitoring package for railway bridges will be marketable in a similar way to the virtual monitoring concept, consisting of two separate parts:
(i) Railway Bridge-WIM system
(ii) Structural health monitoring service
The availability of such a technology will allow the owners of railway bridges to obtain detailed information on the traffic loading being experienced by their bridges, in addition to information on the health of these bridges.

Unique Selling points
During the BridgeMon project the following Unique Selling Points (classified in different categories) were identified:

Identified Technical USPs:
• Class A Bridge-WIM system
o Accuracy levels comparible to, or better than, any other WIM technology.
o Does not require any installation on the road surface.
o Portable – can be used for temporary or short-term monitoring campaigns.
o Can also be used for extended/long-term measurements.
o Provides necessary information to carry out detailed bridge assessments (traffic loading, influence lines, dynamic factors, load distribution, strains etc.).
• Virtual Monitoring
o Provides detailed information on truck weights and traffic flow.
o Detailed modelling of future traffic based on properties of observed traffic.
o Fatigue analysis of all critical locations possible requiring sensors at each location.
• Railway Bridge-WIM/SHM
o Only commercially available Bridge-WIM system for railways.
o Allows accurate evaluation of train weights on a particular railway line.
o Provides information for accurate assessment of the safety of railway bridges.
o Provides information on the response of the structure over time – and allows damage to be identified.

Environmental USPs:
• Reduced requirement for unnecessary replacement or maintenance to bridges due to improved knowledge of bridge safety (i.e. reduction in concrete/steel production).
• Reduced damage to bridges/roads/railway lines as a result of better technologies for monitoring traffic weights and implementing appropriate enforcement procedures for weight limits (reduced damage means less need for carbon-intensive repair/maintenance strategies).
• Reduction in traffic jams arising from unnecessary repairs or maintenace to roads or bridges (traffic jams generate excessive emissions through increased fuel consumption).

Market application and economic USPs:
• Bridge-WIM system is very competitively priced compared to other technologies of the same accuracy.
• No requirement to cut into the pavement during installation.
• No need for road or lane closure during installation.
• Suitable for temporary monitoring on secondary road networks due to the fact that it is not permanent and can be quickly installed and removed.
• Improved accuracy Bridge-WIM can be used as part of an enforcement procedure for legal vehicle weight limits (can be used for preselection of vehicles to be statically weighed by police).
• Virtual monitoring reduces the requirement for installing vast numbers of sensors for detailed monitoring of bridges.
• Accurate projections can be obtained by measuring the traffic for a limited period of time – more accurate information for bridge assessments, road/bridge design, town planning agencies etc.
• Savings to road/rail/bridge owners by avoiding unnecessary replacement and allowing more focused maintenance or repairs to be carried out.

Expected Socio-economic impact of the results
The BridgeMon project has developed a number of technologies which allow advanced and detailed monitoring of bridges and traffic loading to be carried out for both road and railways. Such technologies provide detailed information which ultimately lead to a better understanding of the current traffic loading being experienced on road/rail networks along with the health of bridges on these networks. Every year, bridges are needlessly replaced or rehabilitated – their lives are ended when they still have significant remaining safe working capacity. The developments arising from BridgeMon can be used to target maintenance budgets correctly, avoiding unnecessary rehabilitation and ultimately extending the lives of bridges. In addition, the technologies developed within BridgeMon will also lead to improved design of transport infrastructure. The overall result is a safer, more efficient and more sustainable transport network which is a key component to ensuring the competitiveness of the European and global economy.
The Capacities work programme 2011 (C(2010)4903) emphasises the innovation potential of research and the ever increasing importance of using research to build a knowledge-intensive economy. It identifies the special significance that this takes on in the economic crisis as a means of mitigating its effects and ensuring a rapid recovery in the next economic upswing. BridgeMon is an example of one of the most direct means of translating research into innovation and jobs growth. In this case, the research has directly targeted specific market needs by the SME’s who are working in these markets. These SME’s are at the heart of this project, commissioning exactly the research that they need to gain competitive advantage over international competitors and gain world market share.
Another key impact of the BridgeMon project falls in line with the EU's official policy 'Europe 2020, A European strategy for smart, sustainable and inclusive growth' which highlights the effect of the recession on the growth and job creation achieved within the European Economy in the early part of this century. The policy proposes three priorities to combat this problem: Smart Growth, Sustainable Growth and Inclusive Growth. The outcomes of the BridgeMon project contribute to each of the identified areas.
The BridgeMon project will contribute to Smart Growth by adding ‘intelligence’ to our ageing infrastructure: improved safety through monitoring extends the safe working life of a bridge. BridgeMon delivers market-leading products, giving customers much better information about the safety of their bridges. This is a growth market, especially during a recession, when public budgets are under pressure and bridge owners seek to postpone the construction of replacement bridges.
By extending the safe working lives of existing bridges, BridgeMon improves the
sustainability of transport infrastructure as it reduces the demand for new construction and renovation involving non-renewable resources. Furthermore, renovations of infrastructure cause traffic jams that generate excessive emissions through increase fuel consumption and are not road user friendly. As the technology is implemented by our SME partners throughout the world, this will make a contribution to reducing the demand for concrete and steel worldwide and will demonstrate to others what can be achieved by making our bridges smarter.
Due to the fact that the bridge monitoring process is labour intensive, often requiring local sub-contractors, the technologies developed during BridgeMon will promote long term sustainable and local employment of skilled workers by promoting the advantages of bridge monitoring over the demolish-and-replace alternative. This supports the Inclusive Growth strand of the Europe 2020 policy.

The outputs of the BridgeMon project also contribute to the European growth strategy at a higher level. Europe needs a first class transportation network – road and rail – to keep down the cost of moving goods to market and to promote economic cohesion and a vibrant single market. The single market is at the core of what the European Union stands for. Removing barriers to trade is only part of what is needed to develop and grow the European economy – a world class transportation network is also needed. BridgeMon will make a contribution to the ‘greening’ of transport by reducing the use of non-renewable materials and reducing environmental effects of traffic jams due to road works. Safer, greener and smarter is also the central objective of transport research under FP7.
The Innovation Union is a central part of the Europe 2020 policy, which aims at re-focusing R&D and
innovation policy on the challenges facing our society. BridgeMon has contributed to this through its strong links between public sector research and private sector enterprise and job creation. The project will strengthen existing links and will create new ones. This is in addition to the immediate practical benefits of the research which will result in growth, exports and job creation.

The main socio-economic impact of the BridgeMon technologies can be summarized as follows:
• Strengthening the competitiveness of the SME partners
• Contribution to European Policy
• Job growth
• Government savings on unnecessary repair or rehabilitation of bridges
• Safer transport networks
• More sustainable transport networks
• Reduction in traffic jams due to unnecessary repairs to roads or bridges

Intellectual Property Rights
In the “Research for the Benefit of SMEs” programme, only SME project partners own the results and knowledge developed during the project and shall exploit this Foreground. Effective management of Intellectual Property Rights (IPR) is fundamental in order to derive the greatest commercial value out of the foreground.
Various possible protection routes for IP were disccussed in the consortium: patent, utitlity model, trade secret, knowhow, trade mark.

Until now the consortium did not yet decide to protect any of the identified new foreground, although discussion of this topic was raised at every General Meeting.
Identified exploitable foreground and access rights:
• New Bridge-WIM algorithms
• Virtual Monitoring Concept
• Railway Bridge-WIM system

Conclusions of Business Plan (non-confidential only!)
Much of the world’s bridges are old and need to be periodically repaired or replaced. However, premature replacement wastes money and non-renewable resources. BridgeMon is led by three SMEs who are working to extend the lives of existing bridges. They will develop bridge monitoring technologies to prove that they are safe and can be kept in service for longer, thus ensuring more sustainable road asset management.
A bridge is safe when the stresses due to traffic loads are less than its resistance to these loads. Bridge monitoring includes gathering information about both sides of this inequality, traffic loads and bridge resistance and the BridgeMon SEMs cover both. It is led by Cestel, a world leader in Bridge Weigh-in-Motion (WIM). Cestel have partnered with two SMEs that specialize in monitoring the resistance of bridges to loads, one for the road sector and one for the railway sector. The novel theory of Moving Force Identification (MFI) will be applied in the bridge WIM system to increase accuracy considerably and extend it to other types of bridges. An advance in MFI will be used to extend WIM to railway bridges and to gain additional information on the resistance of bridges to traffic loads. MFI and other improvements will increase the accuracy of the truck weight data to Class A, i.e. to the point where 95% of results have errors of less than 5%, a level not available today with any WIM technology.
The BridgeMon project has generated a new generation Bridge WIM system that will maintain Cestel’s dominant position in the Bridge WIM market for the coming decade and will allow it to gain traction in the general WIM markets. It has extended the Bridge WIM concept from roads into railways and has developed software for the Structural Health Monitoring of railway bridges. This extension brings big opportunity for other BridgeMon partners (Adaptronica, Corner Stone and Kalibra) to generate new flows of income and jobs.
New Bridge WIM concept introduces and demonstrates the simple but highly innovative concept of
Virtual Monitoring of bridges. Together these products will allow the four SME partners of BridgeMon to strengthen and consolidate their positions in their current markets and to extend their product ranges to gain access to new market segments.
In the area of structural health monitoring, BridgeMon will develop and implement tools for improved bridges assessment, including modules to evaluate fatigue life of steel bridges and structural health of railway bridges, both combined with Bridge WIM (traffic loading).
The new technologies, coupled with partnership arrangements with Cestel, will give the other SMEs Unique Selling Points in the bridge resistance monitoring markets. The result gives a potential for massive growth in these SMEs in the coming 5 years, in both turnover and employment.



List of Websites:
bridgemon.zag.si