Bridge Deterioration Models Research Articles

Network-level optimisation approach for bridge interventions scheduling

This paper introduces a novel extension of the multi-system optimisation method, known as the 3C concept, tailored for optimising budget allocation for bridge interventions at the network level. This extended methodology accounts for the interdependencies among bridges due to their spatial proximity within the network. It incorporates direct and user costs, bridge performance indicators, and a bridge deterioration model. A real-world case study involving a portfolio of 555 bridges demonstrates the practicality of the methodology, efficiently determining the optimal intervention sequence. Over an 18-year analysis period, the proposed methodology achieved a 23% reduction in total costs by combining repairs for bridges with high to severe damage and maintenance for the others. This represents a significant improvement compared to the traditional approach, used by bridge management agencies, which relies exclusively on maintenance. The optimised procedure outperforms human intuition in managing complex bridge networks, particularly over extended periods. This methodology can assist transportation agencies in implementing and exploring various scenarios by adjusting the time between consecutive interventions and budget constraints, supporting comprehensive analysis and informed decision-making.

Modelling bridge deterioration using long short-term memory neural networks: a deep learning-based approach

PurposeBridge deterioration is a critical risk to public safety, which mandates regular inspection and maintenance to ensure sustainable transport services. Many models have been developed to aid in understanding deterioration patterns and in planning maintenance actions and fund allocation. This study aims at developing a deep-learning model to predict the deterioration of concrete bridge decks.Design/methodology/approachThree long short-term memory (LSTM) models are formulated to predict the condition rating of bridge decks, namely vanilla LSTM (vLSTM), stacked LSTM (sLSTM), and convolutional neural networks combined with LSTM (CNN-LSTM). The models are developed by utilising the National Bridge Inventory (NBI) datasets spanning from 2001 to 2019 to predict the deck condition ratings in 2021.FindingsResults reveal that all three models have accuracies of 90% and above, with mean squared errors (MSE) between 0.81 and 0.103. Moreover, CNN-LSTM has the best performance, achieving an accuracy of 93%, coefficient of correlation of 0.91, R2 value of 0.83, and MSE of 0.081.Research limitations/implicationsThe study used the NBI bridge inventory databases to develop the bridge deterioration models. Future studies can extend the model to other bridge databases and other applications in the construction industry.Originality/valueThis study provides a detailed and extensive data cleansing process to address the shortcomings in the NBI database. This research presents a framework for implementing artificial intelligence-based models to enhance maintenance planning and a guideline for utilising the NBI or other bridge inventory databases to develop accurate bridge deterioration models. Future studies can extend the model to other bridge databases and other applications in the construction industry.

Development of Data-based Hierarchical Learning Model for Predicting Condition Rating of Bridge Members over Time

Bridge maintenance strategy is implemented to ensure that effective decisions are made based on the evaluation of the current performance and predicted future conditions of a bridge. The current bridge maintenance system relies on traditional manpower-based methods to determine bridge performance. However, the present techniques employ a material deterioration model that has uncertainty in predicting bridge performance. Moreover, the utilization of collected bridge inspection-related data is insignificant, and related research is lacking. Accordingly, a new type of bridge deterioration model is proposed using state performance data based on bridge inspection. To formulate the deterioration model, bridge condition inspection data accumulated over a long period were utilized. The data have been continuously updated since 1992 by the US Federal Highway Administration and consist of basic information and various influencing factors of bridges. The developed model uses the long short-term memory (LSTM) algorithm (a type of recurrent neural network) as well as layer normalization and label smoothing to improve the applicability of fundamental data. For the stable application of data, predicting the model performance for up to 30 years every 2 years is possible. By implementing the many-to-many type of the LSTM algorithm, the predicted probability of each grade derived for each sequence was weighted and averaged with the grade value to derive a continuous state grade result. Thus, the proposed model can express discrete historical degradation indices in continuous form according to the service life. In addition, it enables the prediction of bridge performance using only the fundamental information regarding new and existing bridges. For the effective use of basic data, an optimization model was derived using preprocessing and various regularization techniques. Additionally, a feed-forward deep learning model and stochastic model were developed using the same data. For performance assessment, a regression analysis evaluation method was applied, and comparative analysis was performed using the inspection data of an actual bridge. The use of a time-series model enabled the continuous and stable prediction of bridge performance.

Comparison of Markovian-Based Bridge Deterioration Model Approaches

Comparison of Markovian-Based Bridge Deterioration Model Approaches

Bridge deterioration models for different superstructure types using Markov chains and two-step cluster analysis

Bridges deteriorate due to different causes such as ageing and climatic conditions. As a consequence, repair actions are considered more expensive than maintenance. Therefore, the prediction of maintenance at the bridge network level is particularly complex due to the considerable level of heterogeneity spanning various bridge types and functions. Consequently, in this article an analytical deterioration model called ‘two-step cluster analysis’ is presented and applied to predict the results of a homogeneous data set. This model had been used in different fields including civil engineering but not for superstructure deterioration forecasting. Variables that interfere with the performance of the structure were considered, such as geometry, volume of traffic, among others. The results show that deterioration drop rapidly if repair treatments were not performed. Variations of up to 20.9% at 50 years between the degradation of the database and the analysis of groups, demonstrates the importance of grouping within the analysis of the database, since the late maintenance can affect the life of the bridge. Preventive actions could significantly mitigate the deterioration process. The applied methodology benefits from current bridge inspection practice and produces a probabilistic rating that is consistent with the Markov approach for model deterioration and grouping variables.

Developing Bridge Deterioration Models Using an Artificial Neural Network

The condition of a bridge is critical in quality evaluations and justifying the significant costs incurred by maintaining and repairing bridge infrastructures. Using bridge management systems, the department of transportation in the United States is currently supervising the construction and renovations of thousands of bridges. The inability to obtain funding for the current infrastructures, such that they comply with the requirements identified as part of maintenance, repair, and rehabilitation (MR&R), makes such bridge management systems critical. Bridge management systems facilitate decision making about handling bridge deterioration using an efficient model that accurately predicts bridge condition ratings. The accuracy of this model can facilitate MR&R planning and is used to confirm funds allocated to repair and maintain the bridge network management system. In this study, an artificial neural network (ANN) model is developed to improve the bridge management system (BMS) by improving the prediction accuracy of the deterioration of bridge decks, superstructures, and substructures. A large dataset of historical bridge condition assessment data was used to train and test the proposed ANN models for the deck, superstructure, and substructure components, and the accuracy of these models was 90%, 90%, and 89% on the testing set, respectively.

Analysis of Overweight Truck Permit Policy in New Jersey

The roadway infrastructure constantly deteriorates because of environmental conditions, but other factors such as exposure to heavy trucks exacerbates the rate of deterioration. Therefore, decision-makers are constantly searching for ways to optimize allocation of the limited funds for repair, maintenance, and rehabilitation of New Jersey’s infrastructure. New Jersey legislation requires operators of overweight (OW) trucks to obtain a permit to use the infrastructure. The New Jersey Department of Transportation (NJDOT) issues a variety of permits based on the types of goods carried. These permits allow OW trucks to use the infrastructure either for a single trip or for multiple trips. Therefore, one major concern is whether the permit revenue of the agency can recoup the actual cost of damage to the infrastructure caused by these OW trucks. This study investigates whether NJDOT’s current permit fee program can collect enough revenue to meet the actual cost of damage to the infrastructure caused by these heavy-weight permit trucks. The infrastructure damage is estimated by using pavement and bridge deterioration models and New Jersey permit data from 2013 to 2018 containing vehicle configuration and vehicle route. The analysis indicates that although the cost of infrastructure damage can be recovered for certain permit types, there is room for improvement in the permit program. Moreover, based on permit rules in other states, the overall rank of the New Jersey permit program is evaluated and possible revisions are recommended for future permit policies.

Network-scale deterioration modelling of bridges based on visual inspections and structural attributes

Managing and maintaining bridges on a network-scale is directly associated with the capacity to monitor and forecast the deterioration state of these bridges. Data-driven models such as state-space models (SSM) has been effectively utilized for modelling the deterioration behaviour based on visual inspections of bridge network. However, such a model relies only on the inspection data and does not take into account the structural attributes of each bridge. In addition, the capacity for estimating the deterioration speed is limited, especially in cases with limited number of inspections. In this study, we combine the SSM deterioration model with a kernel regression (KR) method. The SSM-KR framework improves the estimates of the deterioration speed and reduces the overall bias in forecasting the deterioration. The role of KR is to model patterns between the deterioration speed and the structural attributes. Verification and validation of the SSM-KR model are done using synthetic data and real data respectively, whereby the real data is taken from a Canadian bridge network. In addition, the performance of SSM-KR is benchmarked against the existing SSM model using an independent test set of real inspections.

Incorporating the Effects of Climate Change into Bridge Deterioration Modeling: The Case of Slab-on-Girder Highway Bridge Deck Designs across Canada

AbstractClimate change is expected to impact both the operational and structural performance of infrastructure such as buildings, roads, and bridges. However, infrastructure design guides widely re...

Multi-defect modelling of bridge deterioration using truncated inspection records

Bridge Management Systems (BMS) are decision support tools that have gained widespread use across the transportation infrastructure management industry. The Whole Life Cycle Cost (WLCC) modelling in a BMS is typically composed of two main components: a deterioration model and a decision model. An accurate deterioration model is fundamental to any quality decision output.There are examples of deterministic and stochastic models for predictive deterioration modelling in the literature, however the condition of a bridge in these models is considered as an ‘overall’ condition which is either the worst condition or some aggregation of all the defects present. This research proposes a predictive bridge deterioration model which computes deterioration profiles for several distinct deterioration mechanisms on a bridge.The predictive deterioration model is composed of multiple Markov Chains, estimated using a method of maximum likelihood applied to panel data. The data available for all the defects types at each inspection is incomplete. As such, the proposed method considers that only the most significant defects are recorded, and inference is required for the less severe defects. A portfolio of 9726 masonry railway bridges, with an average of 2.47 inspections per bridge, in the United Kingdom is the case study considered.

Deterioration models for prediction of remaining useful life of timber and concrete bridges: A review

Abstract Bridge deterioration models are used for prioritization and maintenance of bridges. These models can be broadly classified as deterministic and stochastic models. There are mechanistic models (or physical models) as well as Artificial Intelligence (AI)-based models, each of which can be stochastic or deterministic in nature. Even though there are several existing deterioration models, state-based stochastic Markov chain-based model is widely employed in bridge management programs. This paper presents a critical review of different bridge deterioration models highlighting the advantages and limitations of each model. The models are applied to some case studies of timber superstructure and concrete bridge decks. Examples are illustrated for arriving at bridge deterioration models using deterministic, stochastic and Artificial Neural Network (ANN)-based models based on National Bridge Inventory (NBI) data. The first example is based on deterministic model and the second on stochastic model. The deterministic model uses the NBI records for the years 1992–2012, while the stochastic model uses the NBI records for one year (2011–2012). The stochastic model is state-based Markov chain model developed using Transition Probability Matrix (TPM) obtained by Percentage Prediction Method (PPM). The two deterioration models (i.e., deterministic and stochastic models) are applied to timber highway bridge superstructure using NBI condition data for bridges in Florida, Georgia, South Carolina and North Carolina. The illustrated examples show that the deterministic model provides higher accuracy in the predicted condition value than the stochastic Markov chain-based model. If the model is developed based on average of transition probabilities considering the data for the period 1992 to 2012, the prediction accuracy of stochastic model will improve. Proper data filtering of condition records aids in improving the accuracy of the deterministic models. The third example illustrates the ANN-based deterioration model for reinforced concrete bridge decks in Florida based on the NBI condition data for the years 1992–2012. The training set accuracy and testing set accuracy in the ANN model are found to be 91% and 88% respectively. The trained model is utilized to generate missing condition data to fill the gaps due to irregular inspections of concrete bridges. This paper also discusses scope for future research on bridge deterioration modeling.

Assessing transition probability of bridge deterioration using Dempster–Shafer theory of evidence

Reliable deterioration models for forecasting the condition of bridge elements, systems, and networks are essential components of any bridge management system. Estimating deterioration transition probabilities is the basis of the popular Markov chain deterioration model. Transition probabilities ha ve been produced in the literature by statistical and probabilistic analysis performed on available bridge inspection and condition data. The available data may be limited or inconsistent with the Markov chain transition period. In addition, the conventional probability theory has limitations when applied to problems with a stochastic nature such as bridge deterioration. This paper introduces a novel method based on the theory of evidence for bridge deterioration modeling through expert judgment elicitation. The advantages of the theory of evidence over the traditional probability theory are discussed and the process for the theory implementation is demonstrated with a case study to validate the application of Dempster–Shafer theory of evidence to estimate the transition probabilities. Based on the results, the theory of evidence is proposed as a scientific expert judgement elicitation technique in the area of bridge condition rating and deterioration modeling. Expert judgment elicitation and theory of evidence application hold potential in the field of bridge management and require further investigation and research.

Development of a Bridge Deterioration Model in a Data-Constrained Environment

AbstractA stochastic time-based deterioration model for use with New Zealand bridges is presented, comprising two parts and being based on the condition management process that is used to assess th...

Framework for Mitigating Human Bias in Selection of Explanatory Variables for Bridge Deterioration Modeling

AbstractThis paper presents a procedure to select explanatory variables, without the influence of human bias, for deterioration model development using publicly available data for bridges in Wyomin...

Performance profiles of metallic bridges subject to coating degradation and atmospheric corrosion

This study focuses on deterioration modelling and performance assessment of metallic bridges affected by atmospheric corrosion, considering also the contribution of typical protective systems in the form of multi-layer coatings. The mechanisms leading to coating degradation are reviewed and the main coating types used by infrastructure owners are highlighted. Building on information contained in industry manuals, a simple model for coating degradation is proposed. Atmospheric corrosion models are then presented, with emphasis given to exposure classification, in line with corrosivity classification guidelines and recent research quantifying the influence of corrosion through dose–response functions. Coating degradation and corrosion models are integrated into a modelling framework, aimed at producing performance profiles of elements in metallic railway bridges. Finally, the framework is implemented in a case study in which a range of condition and resistance performance criteria are presented for different elements, such as girders and stiffeners, and their constituent parts, such as webs and flanges. It is shown that the proposed methodology is sufficiently detailed to enable differentiated performance predictions based on key external factors, and has scope for improvement, especially as coating and corrosion models are informed by the collection of field data.

Modeling Deterioration of Bridge Components with Binary Probit Techniques with Random Effects

Highway agencies use deterioration models to monitor the performance of bridge components (deck, superstructure, and substructure) and to predict their remaining service lives on the basis of the attributes of the bridges and their operating environments. Most bridge deterioration models are deterministic in nature. However, as a result of recent federal legislation, agencies are paying more attention to risk-based performance evaluation and decision making and are seeking models that incorporate stochastic elements. Probabilistic models provide more robust predictions of conditions. This paper describes the development of ordered binary probit (BP) models that duly account for observation-specific effects. The models describe the bridge component deterioration trends, specifically the probability that the component condition will drop from one state to another. This paper also acknowledges past similar or related efforts in this area of research but presents new insights and simplifies the complexity associated with BP models. To demonstrate the application of the models, data from over 5,000 in-service bridges were accessed; included were component age, superstructure material type, type of service under bridge, highway functional class, truck traffic, climate severity, rehabilitation history, condition switching state in last inspection period, and current condition rating. With the use of the developed BP models, a simulation was conducted to predict the probability of the component condition dropping from one state to another, where the predicted future condition is based on simulation involving the predicted probability and the current condition. The paper also presents visualizations of the deterioration trend simulation for each bridge component.

Application of Artificial Intelligence for Bridge Deterioration Model.

The deterministic bridge deterioration model updating problem is well established in bridge management, while the traditional methods and approaches for this problem require manual intervention. An artificial-intelligence-based approach was presented to self-updated parameters of the bridge deterioration model in this paper. When new information and data are collected, a posterior distribution was constructed to describe the integrated result of historical information and the new gained information according to Bayesian theorem, which was used to update model parameters. This AI-based approach is applied to the case of updating parameters of bridge deterioration model, which is the data collected from bridges of 12 districts in Shanghai from 2004 to 2013, and the results showed that it is an accurate, effective, and satisfactory approach to deal with the problem of the parameter updating without manual intervention.

Calibrating Markov Chain–Based Deterioration Models for Predicting Future Conditions of Railway Bridge Elements

Existing nonlinear optimization-based algorithms for estimating Markov transition probability matrix (TPM) in bridge deterioration modeling sometimes fail to find optimum TPM values, and hence lead to invalid future condition prediction. In this study, a Metropolis-Hasting algorithm (MHA)-based Markov chain Monte Carlo (MCMC) simulation technique is proposed to overcome this limitation and calibrate the state-based Markov deterioration models (SBMDM) of railway bridge components. Factors contributing to rail bridge deterioration were identified; inspection data for 1,000 Australian railway bridges over 15 years were reviewed and filtered. The TPMs corresponding to a typical bridge element were estimated using the proposed MCMC simulation method and two other existing methods, namely, regression-based nonlinear optimization (RNO) and Bayesian maximum likelihood (BML). Network-level condition state prediction results obtained from these three approaches were validated using statistical hypothesis tests with a test data set, and performance was compared. Results show that the MCMC-based deterioration model performs better than the other two methods in terms of network-level condition prediction accuracy and capture of model uncertainties.

Implementation of Elman Neural Networks for Enhancing Reliability of Integrated Bridge Deterioration Model

Probabilistic modelling is one of the most prominent techniques in bridge deterioration forecast. It can be classified into two types, namely, state- and time-based models. Reliability of both modelling techniques in forecasting long-term performance rely heavily on sufficient amount of bridge condition rating data being available together with well-distributed deterioration pattern over the age of bridge. However, it can be problematic when the available condition rating records are insufficient. In order to overcome this problem, an integrated deterioration method incorporating both the state- and time-based models has recently been developed. Despite such development and advancement, certain issues still remain with some cases of given condition data that cannot be used to produce reliable long-term performance curve. Aiming to achieve enhanced prediction performance, an Elman neural networks (ENN) technique is incorporated in the integrated method to replace the third-order polynomial regression function, the latter being the core component for long-term prediction in the state-based model. In the present study, the ENN are able to generate more reliable deterioration patterns than a typical deterministic method. The results demonstrate that the integrated method incorporating ENN are more effective in handling various situations of condition data quantities and distributions for generating long-term performance curves.

Implementation of Elman neural networks for enhancing reliability of integrated bridge deterioration model

Probabilistic modelling is one of the most prominent techniques in bridge deterioration forecast. It can be classified into two types, namely, state- and time-based models. Reliability of both modelling techniques in forecasting long-term performance rely heavily on sufficient amount of bridge condition rating data being available together with well-distributed deterioration pattern over the age of bridge. However, it can be problematic when the available condition rating records are insufficient. In order to overcome this problem, an integrated deterioration method incorporating both the state- and time-based models has recently been developed. Despite such development and advancement, certain issues still remain with some cases of given condition data that cannot be used to produce reliable long-term performance curve. Aiming to achieve enhanced prediction performance, an Elman neural networks (ENN) technique is incorporated in the integrated method to replace the third-order polynomial regression function, the latter being the core component for long-term prediction in the state-based model. In the present study, the ENN are able to generate more reliable deterioration patterns than a typical deterministic method. The results demonstrate that the integrated method incorporating ENN are more effective in handling various situations of condition data quantities and distributions for generating long-term performance curves.