BRIDGITISE intends to take benefit from the rapidly growing possibilities offered by new digital technologies to boost the efficiency of the bridge industry.
The overarching research objective is leveraging digital technologies to develop and validate innovative technologies for the cost-effective and sustainable integrity management of bridges.
This research goal presents several technological and economical challenges that must be faced across the value chains of information management and bridge integrity management.
BRIDGITISE’s
overarching research objective is pursued through three Research Clusters (RC) , each playing a pivotal role in the development and validation of cutting-edge technologies aimed at cost-effective and sustainable integrity management of bridges.
RC1
Collect Bridge Information. Innovative devices and technologies will be examined for information collection, such as crowdsensing with smartphones to collect information at a large scale and low cost, InSAR to collect information at a large scale with high precision; vision-based monitoring to increase the accuracy and reliability of weight in motion systems, unmanned vehicles with flight path optimized in real-time based on the collected information; low-cost sensors with edge computing capacity to optimize the volume of information to transmit; and computer vision and domain adaptation to automatize visual inspections of internal bridge areas.
RC2
Process and share bridge information. Advanced technologies to process and share information will be developed. These include BIM integrated within bridge management systems platforms; SHM-informed probabilistic digital twins to support BrIM across lifecycle; cloud platforms to integrate SHM with cybersecurity protocols for the secure storage and sharing of SHM information; neural networks-physic-based hybrid models of deterioration for the prognosis of the bridge performance; and computer vision to automatize the processing of visual inspection information.
RC3
Modeling to support decision. This RC deals with decision problems involved in lifecycle BrIM, such as the development of indicators to maximize the circularity of existing bridges management and the implementation of a framework to estimate the value of digital information. Moreover, the need to update design codes towards a more economical and sustainable design is also considered. Finally, a digital platform enriched with decision analysis modules to optimize bridge lifecycle management will be carried out.
CROWD-Mobile crowdsensing and IoT for bridge system identification
Hosting institution: Politecnico di Milano (POLIMI), Egnatia Odos A.E. (EOD)
Supervisors: M.P. Limongelli (POLIMI), P.F. Giordano (POLIMI), P.Panetsos (EOD)
Objectives: System identification using data collected at a large scale using mobile devices and cloud computing.
Description: Continuous monitoring systems based on local sensors suffer from scalability issues related to the high investments needed to monitor at the network scale. Crowdsensing can significantly reduce large-scale monitoring costs and provide large sets of information, more robust to operational and environmental effects. In the last years, several techniques to extract structural information from sensors mounted on vehicles or smartphones have been proposed. The simultaneous advancements in wireless communication and computing technologies have propelled the use Internet of Things (IoT) and of cloud computing. However, there are still some major research challenges: a) the efficient and reliable transmission of the collected data to a cloud unit for processing; b) the impact of the sensors and carriers (humans and/or vehicles) on the collected data; c) the validation of the approaches on real bridges. The DC will develop a system identification approach based on crowdsensed data. The algorithm will be validated first using the finite element model of a monitored bridge made available by EOD and later on the bridge site. During the second year, the DC will develop an IoT information transmission protocol based on a decentralized architecture, to transmit the crowdsensed data to a platform for processing and sharing. The algorithm will be implemented in a smartphone application to enable the collection and transmission of information by the bridge users. A rewarding game will be ideated to incite the bridge users to use the smartphone application. Results will be compared to those provided by the traditional monitoring system installed on the bridge. The same case study and platform of CYBER and DRONES will be used for data processing.
SATELLITE-Integration of InSAR-derived and environmental measurements for anomaly detection
Hosting institution: Tre Altamira Srl (TRE) Politecnico di Milano (POLIMI)
Supervisors: A. Rucci (TRE), M.P. Limongelli (POLIMI)
Objectives: Detect anomalies in bridges using InSAR, accounting for local environmental and operational effects.
Description: Interferometric Synthetic Aperture Radar (InSAR) monitoring provides millimetric precision information about structural displacements. Their combination with statistical information on other sources of displacement such as weather or traffic, promises to be a key-factor for the detection of anomalies, not yet investigated. SAR techniques have been historically applied to monitor at a large spatial scale. Recent developments in this field enable a higher resolution and cost/benefit ratio compared to traditional topographic survey techniques, for the monitoring of civil infrastructures. In this DC project, the use of InSAR-derived information to detect anomalies is investigated. The objective is to identify signatures of specific damage scenarios for selected categories of bridges in InSAR-derived displacement time histories. To this aim finite element (FE) simulations will be carried out during the secondment at KTH, using the numerical models of real bridges to simulate specified damage scenarios. The numerical responses will be translated into equivalent signals to simulate the noisy measurement that would be acquired by the satellite along the line-of sight (LOS). The labelled LOS equivalent signals will be then used for a twofold aim. Under the supervision of POM a computationally efficient data-driven algorithm based on outliers detection will be developed accounting for the influence of environmental variability. Further to this, during the secondment at KTH and later, under the supervision of TRE, the numerical results will be used to train a supervised Artificial Neural Network (ANN) that will provide, an InSAR-based damage detection tool. The unsupervised and the ANN damage detection tools will be validated using experimental data and the FE-model provided by RAM during short visits of the DC
DRONES-Automated visual inspections of bridges using Unmanned Aerial Systems and vision-based digital imaging
Hosting institution: Politecnico di Bari (POB), Exprivia Spa (EXP)
Supervisors: N. Cordeschi (POB), G. Melone (EXP)
Objectives Optimize Unmanned Aerial Systems-based visual inspections.
Description: The joint use of Unmanned Aerial Systems (UASs) and Artificial Intelligence enables automated visual inspection and assessment through the transmission of real-time collected information to ground control centers, thus greatly facilitating fast assessment of inaccessible areas and/or under hostile environmental conditions, without requiring traffic interruptions. The current state of the art proposes limited applications that do not scale with the employment of multiple UASs. The emerging technologies are only recently facing the challenge of connecting UASs with advanced imaging peripherals relying on algorithms that optimize trajectory and communication quality. Indeed, the integration of UASs with open SHM systems is currently limited due to the lack of state-of-the-art solutions and AI models that can jointly address drone energy efficiency, advanced imaging peripherals, information streaming over the Internet to cloud-integrated services, remote mission design, and real-time control of UASs, that prevent accidents and minimize environmental and telecommunication disturbances factors, e.g., wind, radio interference, low light conditions, optical light reflections, etc. Given the aforementioned technological limitations, this project will develop advanced Artificial Intelligence algorithms for automated visual inspections through the formation of high-speed and high-resolution geo-referenced photos. Furthermore, the control operations and the images are transmitted in real-time through the employment of novel communication techniques and protocols, based on the recent novelties of the 5G communication technologies. The combination of these two approaches leads to a comprehensive framework for bridge assessment and damage detection. Validation of the case study made available by EOD and carried out during a short visit.
WIM-Advanced Bridge Weigh in Motion (B-WIM) performance using vision-based data and machine learning
Hosting institution: Slovenian National Building and Civil Engineering Institute (ZAG), Cestel (CES)
Supervisors: A. Žnidarič, A. Anžlin (ZAG), M. Skobir (CES)
Objectives Leverage vision-based monitoring and machine learning (ML) to enable enforcement-based B-WIM
Description: Prolonging the lifetime of highway bridges based on measured traffic loads is essential for reducing resource exploitation and traffic delays. Financially, this can result in millions of Euros annual savings in maintenance and user costs. Bridge weight in motion systems (B-WIM) can efficiently collect this data, however, their accuracy and reliability do not meet the metrology standard requirements, preventing their use in legal enforcement applications. In this project, a next-generation B-WIM system will be developed leveraging machine learning (ML) technology to process the vast amount of data from different sources (traditional B-WIM, traffic cameras etc.) collected by the industrial supervisor CES. Supported by the expert knowledge of the academic supervisor ZAG, the DC will form data pillars composed of B-WIM measured and processed strain data and synchronized photos acquired by traffic cameras. During the secondment at TWE, a vision-based algorithm will be built to extract vehicle type and dimensions from these photos. Under the supervision of POB, both data pillars will be used to develop a next-level ML classification tool, a prerequisite for a more accurate B-WIM system that will comply with metrology requirements. The ML tool will (i) identify critical aspects in the current B-WIM data processing and (ii), through a sensitivity analysis, recognize data types with the most influential impact on the accuracy and reliability of B-WIM results. Finally, the enriched ML-BWIM will be validated in normal operating conditions under the supervision of CES.
EDGE-Edge computing and dense low-cost sensing for early damage detection
Hosting institution: Sacertis Ingegneria (SAC), KTH Royal Institute of Technology (KTH)
Supervisors: G. Mancini and P. Darò (SAC), R. Karoumi (KTH)
Objectives Develop cost-effective sensors architecture with edge computing capabilities for damage detection.
Description: Damage detection is a key aspect of Structural Health Monitoring (SHM), essential to monitor the time evolution of structural safety. New sensor network architectures, with dense distributions of low-cost sensors, are spreading. The sensor redundancy has the twofold aim to provide a backup in case of sensors nodes failure and to increase the size of the available dataset, thereby reducing uncertainty. Recent developments in edge computing enable the optimization of the sensors’ architecture, in terms of minimum consequences associated with false/missing alarms. In this DC project, these possibilities will be exploited for the development of a Decision Support Tool (DST) for the design of network architecture of edge computing sensors, optimized for damage detection. The main barrier in the development of such DSTs, is usually the lack of real-world data needed for their validation. This DC project will benefit from large datasets made available by SAC. Nonlinear finite element (FE) modelling – calibrated using responses collected on real bridges – will produce large sets of data relevant to several damage scenarios. Using this dataset, a machine learning (ML) algorithm for the classification of the considered damage scenarios will be trained, verified, and validated under the supervision of KTH and used to develop the DST. Sensitivity to non-severe damage, accounting for the metrological performances of the sensors, for the impact of environmental conditions, and the possibility of sensors nodes failures will be investigated to clearly set the limits and benefits provided by the developed DST. The DST will be validated through laboratory testing at POM on a scaled model of reinforced concrete beams and using real data made available by SAC.
ROBOT-Robots, such as Boston Dynamics SPOT
Hosting institution: IBM Research GmbH (IBM) – Swiss Federal Institute of Technology Zurich (ETH)
Supervisors: C. Malossi & F. Scheidegger (IBM), M. Magno (ETH)
Objectives Develop cost-effective sensors architecture with edge computing capabilities for damage detection.
Description: Robots such as SPOT and Unitree A1 have reached outstanding performance in mobility. However, utilization of such devices is still limited in practical visual inspection applications, due to the lack of standardization and automation in the processing of images and videos captured with such devices. Our DC will work towards automating the visual inspection pipeline for these robots to be fully operational. While flying drones are great to inspect external open areas, robots are a better fit for internal areas of bridges, which pose an issue w.r.t. space to fly, presence of dust, and weak GPS signal. At IBM the DC will develop AI algorithms to automate the creation of new domain-specific models using the Robots, as well as automate the domain adaptation process. The DC will study how to enforce real-time quality control of the acquired data, suggest actions to improve the acquisition, automate the selection of relevant frames from long sequence of videos and use self-supervised pipelines to build reusable models with nonannotated data. At ETH the DC will port the AI models on the robot and experiment with multi-modal data collection modalities (e.g., infrared, thermal hyperspectral images). At POM the DC will experiment on real applications supporting data acquisition for the development of the models as well as the validation of the results
PLATFORM- An advanced digital platform to integrate SHM data into bridge management
Hosting institution: Bexel Consulting (BEX), Slovenian National Building and Civil Engineering Institute (ZAG)
Supervisors: S. Lenart (ZAG), I. Osmokrovic (BEX)
Objectives Automatize the integration of continuous monitoring information into Bridge Management Systems (BMS).
Description: BrIM relies on inspection-based performance indicators (PIs) to support decisions on management actions. The integration of information from continuous monitoring systems (SHM), together with the possibility to automatize the estimation of the PIs and facilitate. their visualization through bespoke BIM-based platforms can greatly enhance the support to decision-making these indicators can provide However, such advancements are currently hindered on one side by the unavailability of standardized semantic formats (IFC) to incorporate the SHM information into the BIM model, and on the other by the lack of tools to represent the information conveyed by the PIs in a format that facilitates their use by decision-makers. In this project, a dedicated BrIM extension of a digital BIM platform made available by the industrial supervisor BEX will be developed to enable the integration of a proposed IFC, standard extension incorporating SHM information. Furthermore, a visual representation of the PIs, for the components or selected portions of the bridge, will be enabled based on integrated SHM data into the BIM model. It will be further enriched by a module to estimate the PIs according to the Slovenian BrIM that the DC will develop under the academic supervisor at ZAG and by an algorithm to estimate vibration-based PIs from ambient vibration that the DC will develop during the secondment at POM. The proposed solution will be validated using both a case study made available by ZAG and a selection of bridges in the Tallin local network for which 3D BIM models and data will be made available by TAL.
TWINS-Probabilistic Digital Twins for continuous bridge performance
Hosting institutions: North Consulting (NOR), Sacertis Ingegneria (SAC)
Supervisors: M.H. Faber (NOR), G. Mancini (SAC)
Objectives Develop a probabilistic digital twin model for optimal maintenance management of prestressed concrete bridges.
Description: The multiplicity of possible different and possibly interacting processes contributing to or governing deterioration processes in bridges requires systems representations way beyond the deterministic approach where there is only focus on one or two processes and corresponding models. Currently, there is only limited knowledge of indicators and techniques effective in correlating the observable performance of pre-stressed structures with their structural reliability. A Probabilistic Digital Twin (PDT), supported by SHM and semi-big data from observations and monitoring, can provide a novel contribution to solving this issue and may support integrity management with knowledge improved substantially over time. In this project, an approach to build the PDT of a bridge will be developed making use of nonlinear FE-modelling of the effect of the deterioration. SAC will make available the FE model of a pre-stressed bridge under continuous monitoring that exhibits typical deterioration processes. The DC will use the model to simulate the structural responses under deterioration progressing in time. These responses will be used to develop a PDT model of the bridge that will be utilized as a representation of the best available knowledge and applied to generate (Big Data) simulations of very significant numbers of structural performance scenarios over time; including degradation developments of different origins (environmental factors, loading history, etc.), observations (dynamic responses, displacements, etc.) together with noise associated with different monitoring and inspection devices. Based on the established Big Data, classification schemes will then be utilized to identify how observable structural performances relate to different types and levels of degradation. The PDT will be used during the secondment at POLIMI, as a base to model the structural performances of the bridge under given loading and environment scenarios and identify a procedure to develop the optimal maintenance schedule, based on reliability requirements.
CYBER-Advanced and secure identity provisioning and network monitoring for augmented bridge infrastructures
Hosting institutions: Exprivia (EXP), Politecnico di Bari (POB)
Supervisors: F. Melone (EXP), G. Boggia (POB)
Objectives Design and evaluate a platform for the secure storage and sharing of monitoring data among stakeholders.
Description: High-throughput, real-time, and blockchain technologies enable massive secure communications and data sharing across organizations. Indeed, cybersecurity represents the key enabler for digital chains of trust, ensuring secure communications between all the involved actors in the infrastructure. This allows the application of augmented sensors and communication systems that ensure open SHM systems to operate and collaborate securely. The current state of the art reports several open issues; that limit the broad usage of such security mechanisms, especially in IoT applications. This project will investigate their implementation and integration within an open SHM system, while encompassing authentication and fine-grained access control among peers and information, especially between all the actors involved in bridge monitoring and maintenance. Furthermore, efforts in lightweight cryptography and secure protocols will be researched to securely deliver/receive information to/from constrained devices. To this end, the PhD candidate will (1) study the state of the art in the cybersecurity domain, (2) investigate and validate cybersecurity technology in open SHM systems, (3) design cybersecurity architectures applied to infrastructure monitoring that integrates with an open SHM system, (4) compare conceived solutions with other methodologies through simulations, (5) evaluate the proposed solutions through multiple proofs of concepts, and (6) analyse the performance of conceived solutions and their impact in a real-world SHM scenario using the Greek bridge also used by the DCs CROWD and DRONES.
NEURAL-Machine learning for deterioration prediction based on digital information streams
Hosting institutions: KTH Royal Institute of Technology (KTH), Pedelta (PED)
Supervisors: J. Leander (KTH), J. Jordan (PED)
Objectives Develop prognostic deterioration models based on sensor data and machine learning for service life augmentation.
Description: The project aims at developing a method based on long-term monitoring but with small-scale systems to predict deterioration phenomena such as corrosion wastage and fatigue. The long-term aim is to create a tool for prolonging the service life of bridges and saving condemned structures. Lack of load-bearing capacity shown by theoretical assessments can often be due to a lack of knowledge of the real behaviour of the structure, or overly conservative assumptions. This project will exploit the use of machine learning (ML) to establish metamodels for the interpretation of monitoring data and for the prediction of deterioration, which eventually will lead to a measure of safety (e.g. safety index). Previous research using machine learning has focused typically on damage detection applied to numerical models, simulated damages on laboratory models, and with a few examples on data from real bridges. Examples of data-driven deterioration state prediction for bridges are scarce. Evaluation and validation of several ML algorithms will be conducted in the project to optimize the efficiency of the predictions. The outcome of the project will likely be assembled methods for machine learning combining, e.g., neural networks (NN), support vector machines (SVM), and decision trees (DT), for deterioration prediction. In the second part of the project, the metamodel, developed using numerical data, will be adapted, trained and validated using data collected on real case studies (e.g. HSR Bridge over the Llobregat River, close to Barcelona) under the supervision of PED.
VISUAL-Augmented reality enhanced bridge condition assessment
Hosting institutions: Ingenieursbureau Westenberg B.V. (FWE), University of Twente (TWE)
Supervisors: F. Westenberg (FWE), R. Kromanis, A Hartmann, I. Stipanovic (TWE)
Objectives To automatize the execution of visual inspections and the classification and visualization of damages and defects.
Description: Bridge condition assessment is typically performed through visual inspections that are laborious, and sometimes subjective to an inspector’s opinion. Besides, past inspection reports are seldom used/referred to in repeated inspections. In this project, computer vision, machine and deep learning techniques will be used to build an advanced classification model of bridge elements and defects (extent and severity) from a large database of labelled images made available by the industrial supervisor. The model will then be used to develop an augmented reality (AR) tool to be integrated into handheld devices (e.g., smartphones or tablets) or UASs, to support inspections. The AR tool will allow to (i) indicate areas requiring attention (previously detected defects) and, (ii), reveal propagations of previously detected defects. FWE will provide (i) access to a large database of images of bridge defects with descriptions, (ii) guidance on how to read bridge inspection reports and their metadata, and (iii) define the criteria for defect classification. During the secondment at POB, the DC will develop a ML-based framework to classify visual inspection information and an AR tool that can be used to overlap real-time computed digital content of collected images (e.g., defect type, severity and extend) on the acquired images. TWE will guide the DC in the validation of the classification model and in the application of the AR tool using real case studies made available by FEW.
CIRCULAR-Circular life cycle management of bridges
Hosting institution: University of Twente (TWE), Cemosa (CEM)
Supervisors: I. Stipanovic, A. Hartmann (TWE), N. Jimenez-Redondo, C. Toribio Diaz (CEM)
Objectives To develop circularity metrics as decision-support tools for sustainable bridge management.
Description: Europe has an ambitious plan to become climate neutral by 2050, which can be only achieved by scaling up the circular economy from front-runners to the mainstream economic players, such as construction, and more specifically, the bridge sector. Recycling possibilities for materials, the micro-scale, have been thoroughly researched and predictions for material flow on a city scale, the macro-scale, have been mapped. Yet, the mesoscale, the scale of the construction in general, is scarcely discussed and investigated. Most of the demolished structures are recycled, but downcycled, meaning the material loses value and is primarily used outside the sector. Meso-scale circularity indicators have not been developed yet, because they require to be as detailed as micro-scale indicators yet need to take into account the design principles and associated standardization schemes as well. In this project, the DC will develop circularity metrics and a decision support tool (DST) to link demolishing and new projects exploiting digital data about the existing bridges – including geometry, quality and quantity of the existing materials and components In the first year the DC will perform the analysis of existing Asset Management Platform, will develop mesoscale circularity indicators and metrics for bridges/infrastructure and, using CEM digital platform, will develop different life cycle scenarios (from design to the end-of-life scenario, including recycle, upcycle and reuse). These will be used later to validate the DST developed under the supervision of the academic beneficiary TWE. The DC will perform a secondment at HAC to apply and validate the circularity metrics using data from a Croatian bridge stock. In the second part of the PhD, under the supervision of TWE, the DC will develop a decision support tool for circular bridge management and will apply it to the case studies from Croatia, Spain and Netherlands.
CODES-Small data becoming big data
Hosting institution: Politecnico di Milano (POLIMI), Ramboll (RAM), North Consulting (NOR)
Supervisors: J.H. Roldsgaard (RAM), M.H. Faber (NOR), M. P. Limongelli (POLIMI)
Objectives To develop a methodology to share and integrate monitoring information into bridge design codes
Description: The uncertainty affecting the prediction of the structural behavior and loading is normally represented by safety factors in the design codes. These factors do not account for the availability of information. The access to digital information for bridges has increasingly grown in the last decade. This includes both information gained from sensors, but also inspections, as-built information etc. Their availability has significantly enhanced the modelling of the actions (loading) and their effects for different bridge types. Recently, researchers have been focused upon quantifying the uncertainty of monitoring data to be used in the value quantification. However, bridge-specific or bridge type-specific information is not used in a broader context or in a more generic well-defined context such as how much the potential reduction in safety factor can be based on monitoring data. Design and asset management of a larger group of bridges with similar properties can take advantage of the knowledge gained from the small pool of bridges, where a detailed understanding of the loading and the load effects has been established. This project will develop a framework to (a) codify and publicly share monitoring information collected from a limited pool of bridges of different types. This will include the adoption of robust and reliable online sampling from multiple sources and clients to improve the quality of the data; (b) the extraction of correlations in the monitoring data gathered from a limited pool of different bridges for the creation of a probabilistic prediction model of a bridge to be used in general; (c) the updating of the probabilistic model using monitored data; d) the identification of the process to use the probabilistic model to update design codes in adapted online versions.
CORROSION – Hybrid modelling of corrosion in reinforced concrete structures using heterogeneous data
Hosting institution: Politecnico di Milano (POLIMI), Socotec Monitoring (SMF)
Supervisors: M. Tatin, R. Leclercq (SMF), M.P. Limongelli (POLIMI)
Objectives To use heterogeneous data sources for multiscale predictive corrosion modelling in RC structures.
Description: Corrosion is the main deterioration process affecting reinforced concrete (RC) structures, including bridges. Because of security concerns and maintenance & replacement costs, it has been extensively studied in the past decades. Corrosion involves multiple physical and electrochemical phenomena hence several physics-based models are needed to predict its evolution. Each physics-based model describes one part of the corrosion process e.g., carbonation, chloride ions diffusion, and electrochemical reactions. Besides the lack of comprehensive predictive models of corrosion, approaches to update the model using data from heterogeneous sources are still missing. Usually, data comes from various sources, depending on regulation and on the needs of the asset owner. To fill these gaps, in this individual project, a hybrid modelling approach will be explored combining physics-based models of corrosion with Bayesian Probabilistic Network (BPN) models to integrate data from heterogeneous sources (e.g. visual inspections, core sampling, non-destructive testing (NDT) and embedded sensors). Besides, to optimize the cost-effectiveness of the corrosion monitoring, a module to select the data sources that bring the highest benefits in corrosion maintenance management, a Value of Information (VoI) analysis module will be developed to support the choice of the optimal monitoring system. Because of the cost and time of data collection, in the development phase at POM and LUN the model will be validated using synthetic data. Under the supervision of the industrial supervisor (SMF), the DC will then perform accelerated ageing experiments on RC components and will monitor corrosion with a visual inspection, core sampling, NDT and embedded corrosion sensors. The experimental data will be used for the final validation of the theoretical framework.
VALUE – Value quantification of digital information systems for climate change mitigation
Hosting institution: University of Lund (LUN), Ramboll (RAM)
Supervisors: S. Thöns (LUN), T. Friis (RAM), J.H.Roldsgaard (RAM)
Objectives Develop and validate a framework to quantify the value of digital information to mitigate climate change effects.
Description: The difficulty in estimating the return over the investment in digital technologies makes the end-users (e.g. owners and operators of bridge assets) reluctant to invest in such systems. The possibility to quantify the value of tools for the digital management of information (e.g. monitoring systems, BIM and digital twin models, digital platforms) would have a huge impact on the digitalization of bridge asset management and has been investigated in recent research projects. However, the industrial uptake of such methodologies is quite limited and slow due to the complexity of their application that entails multidisciplinary (measurement engineering, structural engineering, cost-benefit analyses) and transdisciplinary (safety quantification, economic impact) knowledge. A further significant challenge consists in the quantification of the benefit provided by digital information systems when approaching important societal challenges. In this project, the DC will develop a methodological framework for the quantification of the value of digital information systems to support decisions to mitigate climate change effects, namely increased risk of unanticipated hazards (e.g. hydraulic hazards). The benefit of several monitoring methods will be compared using data from ongoing projects and a case study under the supervision of the industrial supervisor (RAM). Consecutively, more efficient and climate-change-risk-reducing digital information strategies will be worked out and demonstrated with a prototype (TRL 7).
D-BIM-Building Information Modelling for decision support
Hosting institution: Wölfel Engineering GmbH (WOL), University of Lund (LUN)
Supervisors: C. Ebert (WOL), S. Thöns (LUN)
Objectives Integration of decision analysis in bridge Building Information Modelling (BIM).
Description: Building information modelling (BIM) and management facilitate the collection and availability of information throughout the life cycle of the built environment. The availability of information by BIM necessitates the integration of the decision modelling to fully benefit from this information as a decision support tool. This DC aims to (1) facilitate the decision analysis and to optimize (cost-effectiveness) the selection of information based on the decision problem at hand; (2) enable the accessibility and availability of the results of decision analysis through digital tools. To this aim, the following tools will be developed in the project (a) models of the life cycle decision scenarios; (b) models of the integrity management procedures, and (c) a building information structure with interfaces to SHM. The interfaces encompass deterioration prediction by machine learning (NEURAL) and measurement data and indicators (DRONES, WIM) as well as digital information systems (VALUE). The enriched BIM model will be validated using case studies made available by BASt, like ZEKISS, BrAssMan, and ROBUST (early warning system with intelligent sensor systems and digital building models, funded by the German Federal Ministry of Education and Research). The validation will lead to the identification of the most efficient integrity management models for relevant decision scenarios and will provide application robustness for the developed prototypic tool (TRL7).