Article
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

The current management process for community buildings in Australia is mainly reactive. Data collected using regular inspections are used for maintenance decision making in the period between consecutive inspections, disregarding the future degradation of the assets and the resultant levels of service. Forecasting future conditions using historic data is difficult because of the uncertainty and stochastic nature of deterioration. A major gap in knowledge is the lack of methods for predicting this highly uncertain degradation process for components of community buildings to support a strategic decision-making process. This paper presents a Markov process-based method for deterioration prediction of building components using condition data collected by the City of Kingston in Australia. Markov transition matrices for building components have been derived using a modified method combining the genetic algorithm with Monte Carlo sampling called direct absolute value difference, which offers superior accuracy. The derived matrices are validated using a new data set collected in 2011. Fourteen transition matrices for building components are proposed. The paper presents a typical decision-making method based on the Markov process.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... This shortfall can be addressed through regular inspection. [6] . ...
... FIGURE 5 shows the example of fragility curves of a physical asset for three scenarios of its aging at 30 years, 60 years and 90 years. The figure shows that with aging, the probability of exceeding the damage level for 5 condition states is increased under same intensity level of damage events FIGURE 5: FRAGILITY CURVES [6] Furthermore, the impacted components might be of different type, for example it can be a physical element such as bridges, rails, rolling stock, or it can be a soft element such as the information and control systems. ...
... The fragility module has to be integrated into the framework after the normal degradation module Mohseni [6] et al. (2017). The input of this module is going to be the initial damage condition and the type and intensity of the disruptive event as shown in FIGURE 6. ...
... Building maintenance is widely acknowledged as a critical issue throughout the building management life cycle to prevent building deterioration [1,2] and to ensure building safety and comfort [3]. Several challenging issues for building maintenance topics, such as maintenance budget [4], demand for safety and serviceability from the organization or user [5], and difficulties in assessing and predicting the future of the facility condition [6], have been receiving great attention in recent years. Especially from the maintenance budget perspective, over a building's life cycle, operation and maintenance costs can account for 50-80% of the total cost [4]. ...
... Building reliable maintenance can be started by establishing a deterioration model to govern decision-making in maintenance management [6,23,25]. Building maintenance actions are complicated by different deterioration rates among various building's functions and locations, as well as coastal conditions, which contribute to more rapid deterioration [23]. Building location also influences other factors that determine deterioration of building facades including concrete formulation, color, texture surface, and finishing surface type [26]. ...
... Equation (5) indicates that if a particular building is selected to be maintained according to a certain maintenance treatment option, the corresponding maintenance cost for the maintenance type maintenance option to upgrade the building condition is also determined. Equation (6) represents that if there is no maintenance treatment applied, the maintenance cost will not occur. Subsequently, the upgrading value from the chosen maintenance treatment is added to the index value condition before maintenance to obtain the index value condition after maintenance. ...
Article
Full-text available
The Indonesian government needs to maintain around 231,000 school buildings in active use. Such a portfolio of buildings given the diversity of locations, limited maintenance budget, and deterioration rates varied by different building conditions presents many challenges to effective maintenance planning. Many of those schools had been reported to be aging and in a degenerated condition. However, contemporary practice for the planning method of Indonesia’s building maintenance program applies reactive maintenance strategies with a single linear deterioration rate. Such methodology cannot properly guarantee the sustainability of those school buildings. Therefore, this study attempts to examine a different approach to Indonesia’s building maintenance planning by adopting a preventive maintenance strategy using the deterioration rate model proved by historical data from a previous study. This study develops an optimization model with varied deterioration rates and considers the budget limitation, by utilizing a Constraint Programming (CP) approach. The proposed model achieves the minimum maintenance cost for a real case of 41 school buildings under different deterioration rates to ensure adequate building conditions and maintain expected levels of service. Finally, research analysis also proves that this new preventive maintenance model has potential to deliver superior capability for assisting building maintenance decisions in Indonesia’s government.
... In a building deterioration prediction context, [28] proposed a Markovian model to estimate the deterioration experienced in community buildings in Australia. They calibrated the transition probability matrices utilized as part of their model by using a nonlinear optimization to minimize the difference between the expected and observed values of building components' condition levels. ...
... In general, according to [28], a Markov chain is a stochastic process allowing for the observation of preset variables at certain points in time, permitting the monitoring of the change in the system states from one time to another. The probability P of a system transitioning from one state m to another state n should conform to the following two conditions: ...
... Eighty percent of the datasets collected were utilized for the design and structuring of the proposed model, while the remaining twenty percent were used for model validation purposes. Two cycles of inspection reports were included in this model as it was suggested by [28] that two is the minimum number of cycles required to derive a reliable and sound Markovian deterioration estimate. After that, uncensored events in the inspection reports were selected as they demonstrate the transition of the system from one state to another and this is the principal tool to build a robust and relatable Markov deterioration prediction model. ...
Article
Full-text available
Healthcare facilities are constantly deteriorating due to tight budgets allocated to the upkeep of building assets. This entails the need for improved deterioration modeling of such buildings in order to enforce a predictive maintenance approach that decreases the unexpected occurrence of failures and the corresponding downtime elapsed to repair or replace the faulty asset components. Currently, hospitals utilize subjective deterioration prediction methodologies that mostly rely on age as the sole indicator of degradation to forecast the useful lives of the building components. Thus, this paper aims at formulating a more efficient stochastic deterioration prediction model that integrates the latest observed condition into the forecasting procedure to overcome the subjectivity and uncertainties associated with the currently employed methods. This is achieved by means of developing a hybrid genetic algorithm-based fuzzy Markovian model that simulates the deterioration process given the scarcity of available data demonstrating the condition assessment and evaluation for such critical facilities. A nonhomogeneous transition probability matrix (TPM) based on fuzzy membership functions representing the condition, age and relative deterioration rate of the hospital systems is utilized to address the inherited uncertainties. The TPM is further calibrated by means of a genetic algorithm to circumvent the drawbacks of the expert-based models. A sensitivity analysis was carried out to analyze the possible changes in the output resulting from predefined modifications to the input parameters in order to ensure the robustness of the model. The performance of the deterioration prediction model developed is then validated through a comparison with a state-of-art stochastic model in contrast to real hospital datasets, and the results obtained from the developed model significantly outperformed the long-established Weibull distribution-based deterioration prediction methodology with mean absolute errors of 1.405 and 9.852, respectively. Therefore, the developed model is expected to assist decision-makers in creating more efficient maintenance programs as well as more data-driven capital renewal plans.
... terioration. Here, the DBN model adopted discrete Markov process that assume a building component's future condition will be dependent on their present condition based on its transition probability (Mohseni et al. 2017). However, in contrast to discrete MC, the transition of moving from a state to another state in BN can be determined from data with different time interval, it can be shorter or longer than one another. ...
... In this study, the component was given condi- tion ratings of C1 ("very good condition") to indicate no damage on the component. Table 13 presents the typical condition rating used in facility management (Mohseni et al. 2017), where Class 1 represents the best condition (very good) and Class 5 represents the worst (very poor) condition. This metric is comparable to seasonal energy efficiency ratio 2 (SEER2) rating system established by the US Department of Energy. ...
Article
Full-text available
The impact of climate conditions on infrastructure is a major concern for the sustainability of built environment. Two main issues that add uncertainty and complexity in climate-change impact are of interest: multiple hazard types and non-stationarity of climate actions. This paper proposes an approach using dynamic Bayesian networks to assess the reliability of a building system considering both gradual and extreme climate factors over the service life of the asset. The methodology is illustrated on a case study that examine an HVAC system, considering overheating fault and degradation risk. Compared to conventional Markov model, the results show stochastic dependence in the degradation process at different time instants and hence affect the variability of degradation. The proposed approach includes economic-based impact analysis to determine costs and payoffs accrued as the consequences. By integrating climate stress and shock and accounting for dynamic changes of the hazard, this method helps decision-makers in identifying and prioritizing adaptation strategies for building system under climate change.
... Therefore, a first step in risk and resilience assessment is to forecast the damage or vulnerability of an asset before the disruptive event takes place. The determination of the future condition of an element or system is a difficult task due to uncertainties and stochastic nature of deterioration (Mohseni, et al. 2017). Several methods exist for damage prediction such as deterministic, statistical and artificial intelligence models. ...
... Finally, the data model is calibrated by means of a direct absolute value difference method with the use of a genetic algorithm for convergence. Details of the methodology can be found in (Mohseni, et al. 2017). ...
Conference Paper
Full-text available
Railway transportation dynamically performs under complex coherent systems and fail-safe interlocking conditions. Security and safety are the first priorities of this massive intermodal transportation. However, railway systems, like any other critical infrastructure, face many threats that can take not only the form of physical, but may also be cyber or combined cyber-physical threats, since the automated-and digital-based technologies in the rail operation may be vulnerable. Therefore, SAFETY4RAILS (S4R), a H2020 EU project, was initiated to strengthen the EU rail operations by increasing the resilience and improving the safety and security of railway and metro networks against these threat types through the further development and combination of a variety of state-of-the-art tools, most of which start with a Technology Readiness Level (TRL) of around 5. Within S4R, many tools will be utilized including a predictive risk and resilience assessment tool, an anomaly detection tool and an asset management tool. To achieve effective tool development and integration, the project requires huge collaboration from different expert and informative sources. This paper will include a discussion on risk and resilience assessment specific to rail systems, including a section on hardware-based countermeasures, before focusing on the risk assessment tool and how it will be implemented. The paper will also introduce a few other tools within the project and discuss the expected interactions the predictive risk assessment tool will have.
... Using this probability, an intervention can be applied, and replacement of the asset can be carried out when a certain probability threshold point is met. FIGURE 13 shows an example of the outcome which is generated by CAMS. Replacement costs are uploaded for each asset. ...
... Markov Chain models can be either timehomogeneous, with constant transition probabilities (the transition matrix is thereby defined as stationary), or time-inhomogeneous, where these probabilities vary over time based on external factors such as environmental changes or maintenance interventions. Expert opinions, bridge type, current condition, environmental factors, and historical maintenance data inform the determination of these transition probabilities [159]. The simplicity and ability to integrate expert knowledge make Markov Chain models popular in BMS applications, such as the implementation detailed in the APTBMS. ...
... In contrast, probabilistic models are based on statistical theory and provide a more realistic approach to predict current and future conditions through a range of possible outcomes [120]. Among these models, Markov chain has been used for the analysis of various infrastructures such as bridges [123], waste water systems [124], stormwater pipes [125], and community buildings [126]. Positive signs of such models are due to the robustness to handle the output of ordinal data type and the probabilistic nature of the underlying deterioration process [122], whereas the models that are sensitive to noisy data [119,127] and the collected data are in subjective nature [122], adding to the list of negatives. ...
Article
Full-text available
Recent developments in networked and smart sensors have significantly changed the way Structural Health Monitoring (SHM) and asset management are being carried out. Since the sensor networks continuously provide real-time data from the structure being monitored, they constitute a more realistic image of the actual status of the structure where the maintenance or repair work can be scheduled based on real requirements. This review is aimed at providing a wealth of knowledge from the working principles of sensors commonly used in SHM, to artificial-intelligence-based digital twin systems used in SHM and proposes a new asset management framework. The way this paper is structured suits researchers and practicing experts both in the fields of sensors as well as in asset management equally.
... CAMS gives an innovative approach to longterm asset management of infrastructure systems (Mohseni 2017). With the understanding of the range of deterioration scenarios for the systems, asset condition data is captured to support risk identification and budget allocation forecasting. ...
Conference Paper
Full-text available
SAFETY4RAILS is the acronym for the European Union Horizon 2020 co-funded innovation project entitled: "Data-based analysis for safety and security protection for detection, prevention, mitigation and response in trans-modal metro and railway networks" which started in October 2020. Its focus is to support the increase of security and resilience against combined cyber-physical threats including natural hazards to railway and metro systems. Its objectives target capabilities to support the characteristics of resilient systems; resilience represented by cycles containing phases of identification, protection, detection, response and recovery (Department of Communications 2019) (or similarly named phases); even if in practice it is not always possible to consider these phases sequentially. An ESREL paper in 2021 offered a very first look into the SAFETY4RAILS project and the SAFETY4RAILS Information System platform as well as some of the tools that are included in the platform. This paper will describe the architectural solution implemented for the platform in the last year and the demonstration of representative capabilities from the first simulation exercise with Madrid Metro at the beginning of 2022.
... CAMS gives an innovative approach to long-term asset management of infrastructure systems (Mohseni 2017). With the understanding of the range of deterioration scenarios for the systems, asset condition data is captured to support risk identification and budget allocation forecasting. ...
Conference Paper
Full-text available
SAFETY4RAILS is the acronym for the European Union Horizon 2020 co-funded innovation project entitled: "Data-based analysis for safety and security protection for detection, prevention, mitigation and response in trans-modal metro and railway networks" which started in October 2020. Its focus is to support the increase of security and resilience against combined cyber-physical threats including natural hazards to railway and metro systems. Its objectives target capabilities to support the characteristics of resilient systems; resilience represented by cycles containing phases of identification, protection, detection, response and recovery (Department of Communications 2019) (or similarly named phases). An ESREL paper in 2021 introduced the SAFETY4RAILS project and the SAFETY4RAILS Information System platform as well as some of the tools that are included in the platform. This paper will describe the architectural solution implemented for the platform in the last year and the demonstration of representative capabilities from the first simulation exercise with Madrid Metro at the beginning of 2022.
... From time to time, they have even been used interchangeably. For example, Sharabah et al. (2007), Edirisinghe et al. (2013), and Mohseni et al. (2017) conducted research on the deterioration of buildings, whereas Doloca (2004), Aisyah et al. (2019), and Ferreira et al. (2021) conducted their research on the degradation of buildings, but based on the same subject matter. Furthermore, the difference between age-based degradation (incurable) and deficiency-based degradation (curable) has not been recognized clearly in the past. ...
... However, records of maintenance or failure of individual building components are not easy to find (Kim et al. 2018). Hence, the usual practice of predicting the deterioration of building assets has been by collecting building condition data based on predefined condition rating systems (Micevski et al. 2002;Mohseni et al. 2017;Edirisinghe et al. 2013)-this is an engineering approach. So if there is, say, a five-condition rating system, each condition would need to be first numbered and labeled (e.g., "like new," "minor repairs required," "should be replaced") and then defined (e.g., "generally discolored," "small cracks," "partially decayed") for each element (e.g., wall painting, wall plaster, timber window). ...
Article
Predictive modeling of deterioration is typically done based on visually inspected condition ratings of building components. This paper presents the concept of a cumulative lost value ratio (CLVR) for component groups to track their deterioration with time. The CLVR is computed through the estimated nominal replacement costs and times for the components that form the group in a given building. This eliminates the need for condition surveys and enables a much larger proportion of building assets to be modeled. Data from the City of Melbourne indicated that eight building component groups encapsulated 87% of building value. The similarities observed in their deterioration patterns were used to allocate them into four categories: (1) superstructure; (2) finishes; (3) fittings, plumbing, and water; and (4) air-conditioning, fire, and electrical. Next, for each year, the proportions of buildings with a given component group in each of five CLVR ranges (C1-C5) were established and fitted with Markov deterioration models. The best fits were obtained for the superstructure and finishes groups and also for Conditions C1 and C5.
... The data-driven decision-making dimension of smart maintenance maturity refers to the level of professionalism in the processing and analyzing of asset data to support decision-making. This dimension addresses the development and use of diagnostic and prognostic algorithms and data science methods that deepen understanding of degradation characteristics and failure probabilities of assets as discussed by Chai et al. (2014), Karim et al. (2016), and Mohseni et al. (2017). While in some situations, machine learning algorithms can trigger dynamic scheduling of maintenance jobs without human interference, in other situations, data has to be interpreted by experts and decisions have to be made by professionals before actions can be scheduled (Lee et al. 2014). ...
... A Markov chain is a random process defined on a discrete state space, characterized by one or multiple dimensions [45,46], for which the conditional probability property holds: Let {x k } be a discrete-time stochastic process that takes its values in space s = {s 1 , s 2 , …s r }. ...
Article
Heavy diesel machinery has a significant contribution to the production of air emissions in large and industrial cities. However, little attention has been paid to this issue, and no appropriate action has been taken to address it. Determining the proper duty cycle can be an effective step in reducing emissions and fuel consumption of these vehicles. The duty cycle includes the driving cycle along with the vehicle operating cycle and makes sense for vehicles that have a specific task. In this article, 80,000 experimental data of the wheel loader (WL) in diggings site by global positioning system (GPS) is collected. After filtering, the data is converted to micro-trips and divided into four clusters using the k-means method, and finally, the Markov matrix is produced. The genetic algorithm (GA) is performed to identify the best combination of micro-trips. In this research, two types of cycles, fuel consumption, and emission were derived using the ADVISOR software. The first cycle is the WL operation at the excavation site, and the second cycle is the short loading cycle. Tehran WL duty cycle has higher fuel consumption, nitrogen oxides (NOx), and hydrocarbon (HC) emission than Environmental Protection Agency (EPA) WL and European non-road transient cycle. While the amount of carbon monoxide (CO) produced in the Tehran WL duty cycle is lower than all EPA duty cycles but higher than the European non-road transient cycle. Tehran WL duty cycle fuel consumption is 265% more than the European non-road transient cycle and 86.12% higher than EPA WL typical operation 1 duty cycle. These differences show that each country’s cycle is only specific to the same country. A derived WL cycle can be used on vehicles with a similar application such as forklift, aircraft support, and forestry equipment.
... Among the categories of deterioration models, Markov chain is a stochastic method for predicting the future condition state of assets in a social infrastructure management system. This method is also most frequently used [30][31][32][33]. Therefore, this study attempts to predict the deterioration pattern of school buildings by using the Markov chain. ...
Article
Full-text available
As the number of aged infrastructures increases every year, a systematic and effective asset management strategy is required. One of the most common analysis methods for preparing an asset management strategy is life cycle cost analysis (LCCA). Most LCCA-related studies have focused on traffic and energy; however, few studies have focused on school buildings. Therefore, an approach should be developed to increase the investment efficiency for the performance improvement of school buildings. Planning and securing budgets for the performance improvement of school building is a complex task that involves various factors, such as current conditions, deterioration behavior and maintenance effect. Therefore, this study proposes a system dynamics (SD) model for the performance improvement of school buildings by using the SD method. In this study, an SD model is used to support efficient decision-making through policy effect analysis, from a macro-perspective, for the performance improvement of school buildings.
... Previous research shows a tendency on integration of sensory systems and database management systems (DBMS) to support facility management activities, based on detection of environmental conditions, room occupancy data, and/or functioning status of plants. Nevertheless, facility management only constitutes a branch, jointly representing real estate market with property management [6,7] and asset management [8] (Figure 1). Asset management, monitoring, and traceability within an organization represent a significant cost element in terms of dedicated resources (times and people) [9]. ...
Article
Full-text available
The integration of facility management and building information modelling (BIM) is an innovative and critical undertaking process to support facility maintenance and management. Even though recent research has proposed various methods and performed an increasing number of case studies, there are still issues of communication processes to be addressed. This paper presents a theoretical framework for digital systems integration of virtual models and smart technologies. Based on the comprehensive analysis of existing technologies for indoor localization, a new workflow is defined and designed, and it is utilized in a practical case study to test the model performance. In the new workflow, a facility management supporting platform is proposed and characterized, featuring indoor positioning systems to allow end users to send geo-referenced reports to central virtual models. In addition, system requirements, information technology (IT) architecture and application procedures are presented. Results show that the integration of end users in the maintenance processes through smart and easy tools can overcome the existing limits of barcode systems and building management systems for failure localization. The proposed framework offers several advantages. First, it allows the identification of every element of an asset including wide physical building elements (walls, floors, etc.) without requiring a prior mapping. Second, the entire cycle of maintenance activities is managed through a unique integrated system including the territorial dimension. Third, data are collected in a standard structure for future uses. Furthermore, the integration of the process in a centralized BIM-GIS (geographical information system) information management system admit a scalable representation of the information supporting facility management processes in terms of assets and supply chain management and monitoring from a spatial perspective.
Article
Systems with dependent main and auxiliary components have been extensively investigated in the reliability field recently, but the influence of the changing environment has been less taken into consideration. Motivated by some real applications, when the protective auxiliary component fails, the degradation/failure rate of the main component varies as it is exposed to different environments. To bridge the gap between research and practice, in this paper the influences of the dynamic environments and the component dependencies are both incorporated to develop a new reliability model for systems with main and auxiliary components. A continuous-time homogeneous Markov process is used to model the evolution of the environments. When the auxiliary component works, it could protect the main component from the negative impact of the environment. Once the auxiliary component fails, the main component would degrade at different rates according to different environment states. Based on the proposed model, first the reliability of the system is derived in a recursive way. Besides, an opportunistic inspection and maintenance policy is designed for the system, and some important indexes such as the limiting average availability and the long-run average cost are derived. Finally, through numerical examples, the applicability of the proposed model and sensitivity analysis of the model parameters are discussed.
Article
Asset management is one of the emerging disciplines in worldwide since most of the organizations are facing challenges in sustaining their operations with the aging of assets. On the other hand, managing of assets and the cost of maintenance have made real-time challenges for them. Hence, the objective of this study is to develop a framework for predictive maintenance of assets considering the existing assessment frameworks of the early researchers. However, the significance of this research is that the framework is developed focusing the passenger terminals where this concept has been narrowly addressed. Kandy Multimodal Transit Terminal was selected for the case study. Accordingly, Markov Chain was used for developing the framework. A conceptual framework was introduced for the predictive maintenance of assets and the cost projections considering the availability of data. Hence, the outcome of this research facilitates to reduce the consequences in asset maintenance management, and to manage the budget allocation for the maintenance in the passenger terminals.
Article
Full-text available
Infrastructural assets such as roads, bridges, and buildings make a considerable contribution to national economies. These assets deteriorate due to aging, environmental conditions, and other external factors. Maintaining the performance of an asset in line with rational repair strategies represents a considerable challenge for decision-makers, who may not pay attention to developing adequate maintenance plans or leave the assets unmaintained. Worldwide, organizations are under pressure to ensure the sustainability of their assets. Such organizations may burden their treasury with random maintenance operations, especially with a limited budget. This research aims to develop a generalized condition assessment approach to monitor and evaluate existing facility elements. The proposed approach represents a methodology to determine the element condition index (CI). The methodology is reinforced with an artificial neural network (ANN) model to predict the element deterioration. The performance of this model was evaluated by comparing the obtained predicted CIs with ordinary least squares (OLS) regression model results to choose the most accurate prediction technique. A case study was applied to a group of wooden doors. The ANN model showed reliable results with R2 values of 0.99, 0.98, and 0.99 for training, cross-validation, and testing sets, respectively. In contrast, the OLS model R2-squared value was 1.00. These results show the high prediction capability of both models with an advantage to the OLS model. Applying this approach to different elements can help decision-makers develop a preventive maintenance schedule and provide the necessary funds.
Article
Operational condition assessments, using a discrete rating system, are frequently used by field engineers to assess inland navigation assets and components. Challenges such as the occasional inability to perform inspections (such as the case with locks watered in an operational state) and protocol requirements requiring ratings even when they aren't inspected lead to highly abstracted inspection data, which are also very prone to human error and misinterpretations due to inspections protocol. On the other hand, some navigational locks are equipped with structural health monitoring (SHM) systems to continuously perform assessments from data obtained in situ. This paper aims to develop a novel hybrid damage prognosis framework for miter gate component of navigational locks, by mitigating effects of human errors on the condition assessment and integrating the highly abstracted inspection data with the SHM. It overcomes two main challenges, namely (1) there is no physical or empirical model available to model the loss-of-contact degradation in the gate, and (2) the mismatches between the inspection data and the SHM system due to data abstraction. A practical case of monitoring loss-of-contact quoin block demonstrates the efficacy of the proposed framework.
Article
Full-text available
While the Markov chain was successfully used for predicting the average condition of a network of assets under one influencing factor, incorporating the effect of a number of factors into a Markov chain model requires a separate analysis of each factor. In particular, modeling building deterioration is overly complicated due to the complexity in number and hierarchy of building components. The research study reported in this paper aims to combine the Markov chain with ISO factor method-based framework to offer a more reliable deterioration forecasting approach for buildings. This paper identifies major influencing factors for building deterioration. It investigates the ability to use Markov chain for deterioration modeling of a selected critical building component. The building condition inspection data from a local government agency in the State of Victoria, Australia, is used to calibrate a Markov deterioration model considering a number of influencing factors. To help demonstrate the concept, two factors influencing deterioration are considered in the analysis: outdoor environment and in-use condition. Validations of the models are undertaken using a subsequent inspection data set using Pearson's chi square test. A generic process for deriving Markov chain-based deterioration curves from discrete condition data incorporating factors influencing building deterioration is presented in this paper. Validated models and the method can be applied to reliably predict the deterioration of buildings with complex hierarchy, which gives a step change in building management.
Article
Full-text available
Multiple constraints imposed by economic, social and environmental considerations lead maintenance planning optimization to a major challenge for designers, owners and users of infrastructure. This study focuses on the management of reinforced concrete (RC) structures placed in chloride-contaminated atmospheres. Under these exposure conditions, chloride penetration generates corrosion of the reinforcing bars reducing the RC durability. Therefore, modeling the deterioration process as well as the maintenance actions carried out during the operational life becomes paramount for the formulation of a comprehensive maintenance strategy. This paper presents a new methodology for improving the performance of maintenance strategies for corroding RC structures. The proposed approach is based on preventive maintenance and combines: a numerical model of chloride penetration, Markov processes and decision theory. The uncertainty related to material properties, model and environmental actions as well as the effect of imperfect inspections are also integrated in the model. The aforementioned points are illustrated with a numerical example. In general, the overall results indicate that the proposed methodology can be useful to improve the costs of a given maintenance strategy by ensuring appropriate levels of serviceability and safety.
Article
Full-text available
In the Mechanistic Empirical Pavement Design Guide (MEPDG), rather than producing a quantitative overall evaluation index, several performance measures are used to separately evaluate the pavement performance condition for rehabilitation design. Due to the characteristics of information on parameters being partially known during the evaluation process, there exist several limitations of the current evaluation methods. In this paper a gray clustering-based methodology is proposed to comprehensively evaluate existing pavements following the MEPDG framework. Gray clustering method is one pillar part of the gray system theory, which was proposed in the 1980s to solve problems with partially known information. The performance criteria used in this study are selected from those defined in MEPDG. The analysis data sets are obtained from the long-term pavement performance (LTPP) database for three types of pavements: flexible pavement, jointed plain concrete pavement, and continuous reinforced concrete pavement. Many LTPP pavement sections are evaluated using the proposed methodology and comparisons are made between gray clustering-based method and extension theory-based approach. It is shown that the proposed gray clustering-based method is able to quantify pavement performance and is suitable to be used as a practical pavement evaluation tool for highway agencies.
Article
Full-text available
Deterioration models are essential components of Infrastructure Management Systems (IMS's) because they predict the future condition of infrastructure facilities and, consequently, assist in optimizing maintenance decisions. Case-Based Reasoning (CBR) is proposed to generate deterioration models that benefit from the large amount of facility data stored in IMS databases and updated on a regular basis. CBRMID (CBR for Modeling Infrastructure Deterioration) is a new CBR system developed to satisfy the special requirements of modeling infrastructure deterioration and to provide government agencies with practical, accurate, and versatile deterioration models. CBRMID is required to support: 1) hierarchical decomposition of infrastructure facilities; 2) representation of facility component interactions; 3) versatility and extensibility of case and knowledge representation; 4) data reusing and sharing; 5) representation of time-dependent data; and 6) fuzziness of retrieval knowledge. In this paper, the architecture of CBRMID is described in terms of case representation, case retrieval, case adaptation, and case accumulation. An application example generated using CBRMID is also presented.
Article
Full-text available
Markovian transition probabilities have been used extensively in the field of infrastructure management, to provide forecasts of facility conditions. However, existing approaches used to estimate these transition probabilities from inspection data are mostly ad hoc and suffer from several statistical limitations. In this paper, econometric methods for the estimation of infrastructure deterioration models and associated transition probabilities from inspection data are presented. The first method is based on the Poisson regression model and follows directly from the Markovian behavior of infrastructure deterioration. The negative binomial regression, a generalization of the Poisson model that relaxes the assumption of equality of mean and variance, is also presented. An empirical case study, using a bridge inspection data set from Indiana, demonstrates the capabilities of the two methods.
Conference Paper
Full-text available
Current annual expenditure for management and renewal of Infrastructure assets around the world is 500 billion US dollars. With an aging stock of infrastructure, innovative methods for management of risk of failure and optimizing of maintenance expenditure becomes extremely important. Whilst different infrastructure assets may have different attributes, governing issues are similar in nature. Prediction of deterioration of some infrastructure is complex since they can constitute of a number of discrete elements with a vast range of influencing factors. A major issue currently faced by local government agencies in Australia is the inability to predict maintenance and replacement expenditure with a reasonable accuracy, which creates situations where emergency repairs would use the funds kept for routine maintenance, which then creates a vicious circle of deterioration.
Book
Recognized as a powerful tool for dealing with uncertainty, Markov modeling can enhance your ability to analyze complex production and service systems. However, most books on Markov chains or decision processes are often either highly theoretical, with few examples, or highly prescriptive, with little justification for the steps of the algorithms used to solve Markov models. Providing a unified treatment of Markov chains and Markov decision processes in a single volume, Markov Chains and Decision Processes for Engineers and Managers supplies a highly detailed description of the construction and solution of Markov models that facilitates their application to diverse processes. Organized around Markov chain structure, the book begins with descriptions of Markov chain states, transitions, structure, and models, and then discusses steady state distributions and passage to a target state in a regular Markov chain. The author treats canonical forms and passage to target states or to classes of target states for reducible Markov chains. He adds an economic dimension by associating rewards with states, thereby linking a Markov chain to a Markov decision process, and then adds decisions to create a Markov decision process, enabling an analyst to choose among alternative Markov chains with rewards so as to maximize expected rewards. An introduction to state reduction and hidden Markov chains rounds out the coverage. In a presentation that balances algorithms and applications, the author provides explanations of the logical relationships that underpin the formulas or algorithms through informal derivations, and devotes considerable attention to the construction of Markov models. He constructs simplified Markov models for a wide assortment of processes such as the weather, gambling, diffusion of gases, a waiting line, inventory, component replacement, machine maintenance, selling a stock, a charge account, a career path, patient flow in a hospital, marketing, and a production line. This treatment helps you harness the power of Markov modeling and apply it to your organization's processes.
Article
The mathematical models investigated are: stepwise regression, B-spline approximation, and constrained least-squares estimation. The best features of each are integrated into an interactive format capable of operating within the PAVER pavement management system. Pavement sections from a given location consisting of the same pavement type, use, and rank are grouped into families. Models that filter obvious errors and statistical outliers from the data are applied to the family data. These procedures constitute a complete method to model and predict pavement family behavior and pavement section behavior accurately.
Article
Timber bridges require high accumulated maintenance costs, which can be many times greater than their initial cost. Infrastructure managers need deterioration models to assist with making appropriate decisions concerning repair strategies and program maintenance schedules by accurately predicting the future condition of timber bridge elements. Markov chain-based models have been used extensively in modeling the deterioration of infrastructure facilities. These models can predict the condition of bridge elements as a probabilistic estimate. This paper presents the prediction of future condition of timber bridge elements using a stochastic Markov chain model. Condition data obtained from the Roads Corporation of Victoria, Australia, were used to develop transition probabilities. The percentage prediction method, regression-based optimization method, and nonlinear optimization technique were applied to predict transition matrices and transient probabilities from the condition data. The most suitable deterioration model for timber bridge elements was selected by evaluating the model performances using the goodness-of-fit and reliability tests. It was concluded that the Markov chain developed for deterioration prediction of timber bridges using the nonlinear optimization technique was mathematically acceptable and predicts the deterioration progression with reasonable accuracy.
Article
Predicting future conditions of pavement plays an important role in pavement management. Prediction for a specific pavement is usually based on the deterioration trend of a group of pavements with similar characteristics, i.e., the same pavement family. This study proposes using the linear mixed effects model (LMEM) to predict future conditions of a specific pavement section by a weighted combination of the average deterioration trend of the family and the past conditions of the specific pavement. The relative weights are determined by the number of past condition measurements available and the degree of variations of the measured past conditions for the specific pavement. The results of the LMEM show significantly higher accuracy in predicting specific pavement conditions compared with two existing adjustment methods that use the last available condition measurement of the specific pavement to adjust the family trend prediction. The finding of this study shows that the LMEM can be used for project level pavement condition prediction or other types of infrastructure condition prediction, whereas future conditions of a specific entity are to be projected based on a combination of the average "family" trend, as well as the individual's condition history.
Article
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.
Article
Buildings are one of the major infrastructure investments in cities. Sustainable preservation of building assets in order to deliver an appropriate level of service throughout their life cycle requires a comprehensive and optimised decision making methodology. This decision making method needs to be supported not only by accurate data, but also by proper manipulation and aggregation techniques to target the highest potential longevity of construction materials. Condition monitoring methods help asset managers collect required information about their buildings to make justifiable judgments for maintenance and rehabilitation strategies. This data is collected in the condition monitoring stage within a defined scope of a condition monitoring manual. The level of detail in data collection may depend on the asset management system, element hierarchy adopted by the organization and criticality of assets. While detailed condition data is collected during building condition assessments, for higher-level optimised strategic asset management overall conditions of element groups are desirable to project the capital investments and expenditures. The paper reviews condition monitoring techniques for buildings and also presents a risk-based methodology for aggregating the inspected conditions to a higher group level of inspected elements which leads to a greater accuracy of decisions to be made for strategic management of buildings.
Article
A mathematical model was developed to represent pyrolysis. The components of primary and secondary pyrolysis reactions were simply lump into different groups and were represented through a set of pseudo-first-order reactions. This study presents an algorithm to estimate the kinetic parameters using Monte-Carlo (MC) simulation. The combination of an analytical reaction model and the MC simulation technique rapidly generates a large number of numerical values. Results show that MC-simulated data and experimental data are in fair agreement. Though the technique developed in this study proved to have potential, more experimental data are needed to check the robustness of the model. © 2011 Canadian Society for Chemical Engineering
Article
Effective management of storm water pipe networks requires an accurate assessment of the structural condition of the pipes. Indeed, the recent introduction of Australian accounting standard AAS27 compels local governments to prepare annual financial statements, including the depreciated value of their storm water network. A rational approach to assessing depreciation is to base it on structural deterioration. This study presents a Markov model for the structural deterioration of storm water pipes. The model is calibrated, using Bayesian techniques, to structural condition data from the storm water asset database of the Newcastle City Council (Australia). It is shown that the Markov model is consistent with the data. The pipe characteristics of diameter, construction material, soil type, and exposure classification were found to influence the deterioration process. It is also shown that the depreciation methods required by AAS27 significantly overestimate the structural deterioration.
Article
The chemical parameters used in reactive transport models are not known accurately due to the complexity and the heterogeneous conditions of a real domain. The development of an efficient algorithm in order to estimate the chemical parameters using Monte-Carlo method is presented. By fitting the results obtained from the model with the experimental curves obtained with various experimental conditions, the problem of parameters estimation is converted into a minimization problem. Monte-Carlo methods are very robust for the optimization of the highly nonlinear mathematical model describing reactive transport. It involves generating random values of parameters and finding the best set. The focus is to develop an optimization algorithm which uses less number of realizations so as to reduce the CPU time. Reactive transport of TBT through natural quartz sand at seven different pHs is taken as the test case. Our algorithm will be used to estimate the chemical parameters of the sorption of TBT onto the natural quartz sand. © 2006 American Institute of Chemical Engineers AIChE J, 2006
Article
Maximum likelihood fit of nonlinear, implicit, multiple-response models to data containing normally distributed random errors can be carried out by a combination of the Gauss-Newton generalized nonlinear least-square algorithm first described by Britt and Luecke in 1973, with a Fletcher-Reeves conjugate gradient search for initial parameter estimates. The convergence of the algorithm is further improved by adding a step-limiting procedure that ensures a reduction in the objective function for each iteration. Multiple-equation regression methods appropriate to the solution of explicit fixed-regressor models are derived from this general treatment as special cases. These include weighted nonlinear least squares (where the covariance matrix of the response is known), and uniformly weighted nonlinear least squares (where the responses are uncorrelated and characterized by a single common variance). Alternative methods for fixed-regressor fits of explicit multiequation models with an unknown covariance matrix of the responses are also considered. The moment-matrix determinant criterion appropriate in such situations is also efficiently minimized by use of the conjugate-gradient algorithm, which is considerably less sensitive to the accuracy of the initial parameter estimate than the more usual Gauss-Newton methods. The performance of the new algorithm for models defined by one, two, and three implicit functional constraints per point is illustrated by random-regressor fits of isothermal p−X and p−X−Y vapor–liquid equilibrium data, and ternary liquid–liquid equilibrium data, respectively.
Article
An adaptive parameter estimation algorithm for a class of biochemical processes expressed by a nonlinearly parametrized Monod's growth kinetics model is presented. Contrary to conventional least-square or gradient-type identification techniques, the proposed parameter estimation algorithm is developed based on Lyapunov's stability theory. A novel class of parameter-dependent Lyapunov functions is utilized to remove the difficulty associated with estimating the unknown parameters that appear nonlinearly. A persistence of excitation (PE) condition is investigated to guarantee the convergence of the estimation scheme. Simulations are provided to verify the effectiveness of the new approach and the theoretical discussion.
Article
In comparison with the well-researched field of analysis and design of structural systems, the life-cycle performance prediction of these systems under no maintenance as well as under various maintenance scenarios is far more complex, and is a rapidly emergent field in structural engineering. As structures become older and maintenance costs become higher, different agencies and administrations in charge of civil infrastructure systems are facing challenges related to the implementation of structure maintenance and management systems based on life-cycle cost considerations. This article reviews the research to date related to probabilistic models for maintaining and optimizing the life-cycle performance of deteriorating structures and formulates future directions in this field.
Book
From the Publisher: A groundbreaking addition to the existing literature, Exploratory Data Mining and Data Cleaning serves as an important reference for data analysts who need to analyze large amounts of unfamiliar data, operations managers, and students in undergraduate or graduate-level courses, dealing with data analysis and data mining.
Article
Remaining useful life (RUL) is the useful life left on an asset at a particular time of operation. Its estimation is central to condition based maintenance and prognostics and health management. RUL is typically random and unknown, and as such it must be estimated from available sources of information such as the information obtained in condition and health monitoring. The research on how to best estimate the RUL has gained popularity recently due to the rapid advances in condition and health monitoring techniques. However, due to its complicated relationship with observable health information, there is no such best approach which can be used universally to achieve the best estimate. As such this paper reviews the recent modeling developments for estimating the RUL. The review is centred on statistical data driven approaches which rely only on available past observed data and statistical models. The approaches are classified into two broad types of models, that is, models that rely on directly observed state information of the asset, and those do not. We systematically review the models and approaches reported in the literature and finally highlight future research challenges.
Article
Like other engineering structures, buried stormwater drainage pipes deteriorate and fail over time in terms of pipe collapses due to structural deterioration or pipe blockages due to hydraulic deterioration. The deterioration of service infrastructure was a concern in Australian in recent times, where stormwater drainage pipes in Australia were rated as ‘poor condition’. The information on current and future condition of stormwater pipes is therefore important for making decisions on when and how to carry out maintenance and rehabilitation. As the major objective, this study attempted to develop separately structural and hydraulic deterioration models that can predict the condition changes of pipe population and condition changes of individual pipes as compared to the ‘like-new’ condition. The outcomes of the models can be used for planning annual budget and prioritizing repairs. Furthermore, this study aimed to identify the significant factors that affect the structural and hydraulic condition of stormwater pipes, which could support design and operation of stormwater pipes. To achieve these objectives, this study first considered an ideal deterioration model which recognized that pipes deteriorate differently due to their contributing factors such as pipe size and soil type. Based on the ideal deterioration model, five practical deterioration models were developed using statistical techniques and neural networks (NNs), and were calibrated using different optimization techniques in searching for the best suitable model. These deterioration models were selected considering the availability of snap-shot (or once only) inspection data and the ordinal grading system of pipe condition. The model inputs were contributing factors and the model output was pipe condition in ordinal numbers. Methods for assessing the predictive performance of these models and determining the significant input factors were considered. A case study with data collected from a City Council in Melbourne (Australia) was used to demonstrate the applicability of the models developed in this study. The results showed that the NN model and the Markov (statistical) model were the best models for predicting condition changes of individual pipes and pipe population respectively. Several factors such as pipe size and pipe location were found significant factors in these models. The significance of this study is the development of deterioration models that provide a basis for the construction of a comprehensive asset management system for stormwater pipes. The major innovation of this study is the exploitation of advanced modelling techniques for predicting the deterioration process of stormwater pipes.
Article
Presented at the 10DBMC International Conférence On Durability of Building Materials and Components, Lyon, France, 17-20 April 2005 Making decisions on building maintenance policies is an important topic in facility management. To evaluate different maintenance policies and make rational selection, both performance and maintenance cost of building components need to be of concern. For roofing sytem Markov Chain model has been developed to simulate the stochastic degrading process to evaluate the life cycle perfornance and cost. [Van Winden and Dekker 1998; Lounis et al. 1999] Taking value in a discrete state space, this model is especially appropriate when scaled rating regular inspections and related mainteance policies are implemented in large organizations. [Van Winden and Dekker 1998] However, many parameters in this Markov Chain model are associated with variance of significant magnitude. The propagation of these variances through the model will result in uncertainties in predicted life cycle performance and cost results. Without a solid uncertainty analysis on the simulation, decisions based on these simulation results can be unrealiable. In this paper we provide methods to estimate the range of parameter values and represent them in a probabilistic framwork. Monte Carlo method is used to analyze simulation output (life cycle cost and performance) variance propagated from these parameters through the model. These probablisitc informnation can be used to make better informed decisions. An example is provided to illustrate the Markov Chain model development, parameter identification method, Monte-Carlo uncertainty assessment and decision making with probabilistic information. It is shown that the uncertainty propagating through this process is not negligible and may significantly influence or even change the final decision
A reliability based approach for service life modeling of council owned infrastructure assets
  • A Sharabah
  • S Setunge
  • P Zeephongsekul