Conference PaperPDF Available

Uncertainty in condition prediction of bridges based on assessment method – case study in Estonia

Authors:

Abstract

In this paper the uncertainty in condition assessment based on most common assessment methods, visual inspection and non-destructive testing, is investigated. For decision-making the averaged or estimated value is suitable, but if the basis of a decision is only a subjective visual inspection, then it could lead to a wrong decision. The second most traditional assessment method is non-destructive testing (NDT), which can give reliable results, but the interpretation of measurement is needed. To investigate the errors in both evaluations, benchmarking tests were carried out in Estonia within two groups, a group of experienced inspectors and a group of unexperienced students, to show how the importance of experience affects results. To present the influence of assessment uncertainty to condition prediction curves based on continuous-time Markov model are calculated and for updating, Bayesian inference procedure is used.
Conference Paper
p>Railway bridges represent crucial points within railways, because of its strategy meanings in transportation infrastructure. During the decades of exploitation, degradation processes and external influences attack the bridge structures. Typical 70-years old riveted railway bridge with main plate girders and the upper member deck is analysed in the paper. Load-carrying capacity, possible reserves and analysis of critical details are presented. Provisions of the new guideline for determination of load-carrying capacities of railway bridges were applied. Data needed for numerical analysis and subsequent calculation were supported by enhanced diagnostics and measurements. Actual behaviour of the whole structure and chosen members, respectively, were verified by the static and the dynamic load tests. Both the researches academics from the university and the R&D institute staff of the state railway administrator were involved in this pilot project.</p
Article
Full-text available
This paper introduces a possible way to use a multivariate methodology, called principal component analysis, to reduce the dimensionality of condition state database of bridge elements, collected during visual inspections. Attention is paid to the condition assessment of bridges in Estonian national roads and collected data, which plays an important role in the selection of correct statistical technique and obtaining reliable results. Additionally, detailed overview of typical road bridges and examples of collected information is provided. Statistical analysis is carried out by most natural reinforced concrete bridges in Estonia and comparison is made among different typologies. The introduced multivariate technique algorithms are presented and collated in two different formulations, with contrast on unevenness in variables and taking into account the missing data. Principal components and weighing factors, which are calculated for bridges with different typology, also have differences in results and element groups where variation is retained.
Article
To assess concrete strength in a structure, nondestructive technique (NDT) like rebound hammer is combined with destructive technique (coring tests) in order to implement a relationship ‘‘conversion model” between the compressive strength and, NDT measurements. The conversion model is used to estimate the local strength value at each test location using the corresponding NDT value. Then the estimated mean strength and/or estimated strength standard deviation (concrete strength variability) values are calculated. However, the reliability of these estimated values is always a questionable issue because of the uncertainties associated with the strength predictions based upon NDT measurements. To improve the reliability, the uncertainties must be reduced by specifying and controlling their influencing factors. The objective of this paper is to study the reliability of assessment by analyzing the effects of several influencing factors: number of test locations used to identify a conversion model between strength and rebound measurement NC (number of cores), true value of concrete strength variability, within-test variability of rebound measurements, accepted uncertainty level, quantity to be assessed (mean strength, strength variability), model identification approach (like regression) and the way of selection core locations (random or conditional i.e. selection based on NDT measurements from preliminary investigation). To this end, a large campaign of laboratory studies datasets (1700 test results) was considered for the analysis in the present study.
Article
During operation, bridge condition is decreasing due to various effects. The condition is described by "condition rating coefficient", determined by visual assessment. Condition of Slovenian state road bridge network is analyzed. The condition rating database is structured with respect to traffic load, climate zone, structural material, structural type and bridge element. Changes of condition rating coefficient with time are presented and analyzed. Results show that climate and exposure to water are the most important parameters influencing the condition of bridges. Challenges related to assessment methodology are identified and discussed.
Book
This book examines advanced Bayesian computational methods. It presents methods for sampling from posterior distributions and discusses how to compute posterior quantities of interest using Markov chain Monte Carlo (MCMC) samples. This book examines each of these issues in detail and heavily focuses on computing various posterior quantities of interest from a given MCMC sample. Several topics are addressed, including techniques for MCMC sampling, Monte Carlo methods for estimation of posterior quantities, improving simulation accuracy, marginal posterior density estimation, estimation of normalizing constants, constrained parameter problems, highest posterior density interval calculations, computation of posterior modes, and posterior computations for proportional hazards models and Dirichlet process models. The authors also discuss computions involving model comparisons, including both nested and non-nested models, marginal likelihood methods, ratios of normalizing constants, Bayes factors, the Savage-Dickey density ratio, Stochastic Search Variable Selection, Bayesian Model Averaging, the reverse jump algorithm, and model adequacy using predictive and latent residual approaches. The book presents an equal mixture of theory and applications involving real data. The book is intended as a graduate textbook or a reference book for a one semester course at the advanced masters or Ph.D. level. It would also serve as a useful reference book for applied or theoretical researchers as well as practitioners. Ming-Hui Chen is Associate Professor of Mathematical Sciences at Worcester Polytechnic Institute, Qu-Man Shao is Assistant Professor of Mathematics at the University of Oregon. Joseph G. Ibrahim is Associate Professor of Biostatistics at the Harvard School of Public Health and Dana-Farber Cancer Institute.
Conference Paper
Managers of civil infrastructure aim at reaching the best balance between life-cycle performance of structures and the necessary maintenance, repair and replacement costs. However, long time prediction of performance, considering the effects of deterioration and maintenance actions, can not be very accurate, and decisions must be made considering the uncertainties in initial performance, deterioration, and time of application and effects of maintenance actions. The uncertainties in performance can be reduced through inspection, non-destructive testing or load tests. However, information from these sources is not completely accurate and must be combined with existing information, in order to obtain reliable posterior information and predict more accurately future deterioration. In this paper, the application of maintenance actions on existing bridges is optimized in order to obtain maximum performance and minimum life-cycle maintenance cost. Performance is defined in terms of the condition index, which describes the effects of deterioration as can be seen by an inspector, and the safety index, which measures the safety margin of the structure. Constraints are considered for both the condition and the safety index over the entire lifetime. Several maintenance actions are considered, each defined by probabilistic effects on performance, times of application and costs of applications. The reduction in uncertainty associated with periodical inspections is considered through updating of performance profiles. Considering multiple objective functions as well as the complexity of the performance profiles under maintenance, genetic algorithms are combined with Monte-Carlo simulation, in order to obtain a set of Pareto solutions. Examples of application are presented, showing the effects of inspection and updating on the optimization of maintenance strategies.
Article
There should be no argument that the quality of life and the economic vitality of modern civilizations are dependent on the infrastructures that support them. Of major importance is the transportation system that provides for the transport of goods and services as well as the mobility of the citizens within those civilizations. The enormous investment that Governments have made in highway systems since the 1950's has led the U.S. to be recognized as having the largest and most modern highway system in the world. Within this highway system, bridges are singly the most important link. Without bridges, highway transportation as we know it would cease to exist. In recognition of this, the California Department of Transportation continues to take aggressive actions in the protection of its bridge infrastructure as evidenced by the delivery efforts of the Seismic Retrofit Program and the creation of the Bridge Scour Mitigation Program.
Article
The focus of this paper is on the statistical estimation of parameters in various types of continuous-time Markov processes using bridge condition data in the Netherlands. The parameters in these processes are transition intensities between discrete condition states. These intensities may depend on the current state and on the age of the structure. The likelihood function for the observed transitions is derived and an outline of the steps to determine the parameters which maximize the likelihood function is given. The results of the fitting procedure are presented in the form of the expected bridge condition over time and the random time to the final state. For each type of Markov process, the quality of fit is compared to that of the other types. The influence of the inspector on the condition process is also discussed.