Use Of Bayesian Belief Networks Research Articles

The use of Bayesian Belief Networks (BBNs) to probe deeper into railway safety management systems – Two studies from Great Britain and Italy

The importance of Safety Management Systems (SMS) to the railway industry is underlined by the fact that all organisations operating on UK railways are required by law to have one. Analysing SMSs can provide a reliable systemic tool to identify hazards and weaknesses within complex systems like the railway, making it possible to significantly increase safety, reducing the odds of near misses and accidents. However, there is little empirical research evidence to determine the impact on safety of a structured SMS. The current paper describes two studies which use Bayesian Belief Networks (BBN) to conceptualise SMSs and their impact on front-line performance. The paper presents the usefulness of BBNs to compare complex systems and reconcile cultural differences within the railway industry, identifying factors that are deemed vital within Italy and Britain. The two studies allowed us to identify the most influential factors within a SMS and how they interact with each other, as well as the strength of the identified relationships. A BBN is particularly useful in estimating how changing some of the node states (e.g., by making safety leadership present) affected the other factors. The current study showed that safety leadership has an impact on the SMSs of the British and Italian railway industries.

Integrating Bayesian Belief Networks in a toolbox for decision support on plastic clean-up technologies in rivers and estuaries

Current mitigation strategies to offset marine plastic pollution, a global concern, typically rely on preventing floating debris from reaching coastal ecosystems. Specifically, clean-up technologies are designed to collect plastics by removing debris from the aquatic environment such as rivers and estuaries. However, to date, there is little published data on their potential impact on riverine and estuarine organisms and ecosystems. Multiple parameters might play a role in the chances of biota and organic debris being unintentionally caught within a mechanical clean-up system, but their exact contribution to a potential impact is unknown. Here, we identified four clusters of parameters that can potentially determine the bycatch: (i) the environmental conditions in which the clean-up system is deployed, (ii) the traits of the biota the system interacts with, (iii) the traits of plastic items present in the system, and, (iv) the design and operation of the clean-up mechanism itself. To efficiently quantify and assess the influence of each of the clusters on bycatch, we suggest the use of transparent and objective tools. In particular, we discuss the use of Bayesian Belief Networks (BBNs) as a promising probabilistic modelling method for an evidence-based trade-off between removal efficiency and bycatch. We argue that BBN probabilistic models are a valuable tool to assist stakeholders, prior to the deployment of any clean-up technology, in selecting the best-suited mechanism to collect floating plastic debris while managing potential adverse effects on the ecosystem.

Dynamic Bayesian Belief Network for long-term monitoring and system barrier failure analysis: Decommissioned wells

There is increasing interest to consider dependent failures and human errors in the offshore industry. Permanently abandoned wells dot most of the subsea environment. The nature of a well plugging and abandonment (Well P&A) run - usually the lowest-cost contractor engaged to plug several wells tapping the same reservoir makes it an ideal case study for incorporating failures based on common causes. The heavy use of operators during a cementing job also provides the case for analysis of human error in such tasks. One proposed method to analyse the above-mentioned is the use of Bayesian Belief Networks to achieve the following objectives (1) to capture better estimates of a well PA event by incorporating dependencies, and meet regulatory requirements by authorities; and (2) to use the same model to provide long term monitoring of a group of wells linked by common dependencies. This model has not only captured the dependencies of multiple variables, but also projected it in a dynamic manner to provide a risk profile for the next decade where well integrity failure is likely to happen.• Proposed adapted method capture better estimates of a well PA event by incorporating dependencies• Method allows for extension of model to long term monitoring of a group of wells linked by common dependencies

A 20‐year mapping of Bayesian belief networks in software project management

In recent years, more software engineering researchers have focussed on methods, techniques, tools, and processes to support software project management (SPM) addressing quality, cost, and time constraints. In this direction, the use of Bayesian belief networks (BBNs) has gained attention for providing a powerful mechanism for cause–effect analysis with both qualitative and quantitative data to support decision making under uncertainty. This work aims to provide an overview of the first 20 years of research in BBNs applied to SPM, and by so doing it contributes to the structuring of this research topic and the identification of new research opportunities. We conducted a systematic mapping (SM) study on the use of BBN as a decision-support tool for SPM issues. We analysed over 109 relevant publications, from 1999 to 2018 (i.e., 20 years), to understand the motivations for using Bayesian networks (why), the problem domain and model scope addressed by researchers (what), the stage of the life cycle in which BBNs are being used (when), the venues of the publications (where), and the tools used to model the Bayesian networks (how). We draw the following conclusions from the results of our SM: (1) The application of Bayesian networks in SPM has been an active research topic since 1999. (2) Prediction and planning are the most common purposes in 60% of selected papers. (3) Software quality (55%) is the problem domain most investigated. (4) Most of the surveyed studies embrace the project (47%) and product (37%) scopes. And finally, (5) the development phase (86%) is when Bayesian networks have been more used. As future work, we highlight the need to further investigate methodologies that enable the use of BBNs in real software development contexts.

Towards understanding the role of islandness in shaping socio-ecological systems on SIDS: The socio-ecological islandscape concept

Small Island Developing States are under significant threat from global change and unsustainable local practices. These issues are generally exacerbated by an inherent vulnerability to social, economic and ecological changes consequential to their unique island characteristics (e.g. limited resources), which is often described as “islandness”. These unique island characteristics or “islandness” have shaped, and continue to shape, their socio-ecological systems which, from a real-world geographic perspective, the island and its communities form an integrated socio-ecological landscape. It can be argued that “islandness” contributes immensely to understanding the socio-ecological systems of an island, and in considering “islandness” as a fundamental element that significantly influences socio-ecological systems on islands, socio-ecological systems then materialise as socio-ecological islandscapes. This study ultimately sets out to initiate the development of the socio-ecological islandscape concept through (1) providing the preliminary framework, and (2) gaining insights into what impact each of the unique characteristics of “islandness” has on a system, both as individual components and as a collective, through the use of Bayesian Belief Networks. The results of the study provide evidence of the influence that “islandness” has on both the social and ecological components of a socio-ecological system. Differences in degree of influence between the unique island characteristics (e.g. size vs. isolation) also showed how some characteristics have greater impacts. The socio-ecological islandscape framework proposed in this study propagates a research agenda focused on the SES-landscape-SIDS-islandness nexus, due to its apparent effect on the economic disadvantages and ecological fragility faced by SIDS that threatens their sustainability.

Score-based Bayesian belief network structure learning in damage risk modelling of mining areas building development

The article describes the results of using the application of Bayesian approach, to create a decision support system to evaluate the risk of the damages to prefabricated reinforced concrete (RC) buildings, that have been exposed to the impacts from the industrial environment of mines in the form of spatial subsidence and tremors. This problem is important from the point of view of both structural safety and utility aspects. It concerns many buildings, in which the damage is often caused by the excessive consumption of thermal energy used for heating interiors or dehumidifying damp walls. The above issues make the subject extremely important from a technical, socio-economical, and energy point of view, therefore it can be categorised as sustainable development problem. The Bayesian belief network (BBN) is used to build a risk model that encompasses all the issues and variables that describe the damage process associated with the entire scope.The use of BBNs is an innovative approach in this interdisciplinary area, that encompasses civil engineering and mining environmental protection. There are two types of score-based methods for learning the structure of a BBN from data which will be applied and compared. The first is an iterative method for searching the space of potential solutions, based on the Tabu-search algorithm. The second approach used a method of stochastic searching for a space of structures (node orders) based on the global optimisation algorithm Simulated Annealing (SA). The analyses were performed in R, using the catnet and bnlearn packages. The basis for the study was a database relating to 129 prefabricated reinforced concrete buildings with a wall bearing structure, located in a copper mining area in Poland. Finally, there were eight damage intensity indexes (wui) obtained from using the above models when analysing the risk of damage to structural and finishing elements. The most effective method turned out to be using the Tabu-search algorithm with the assumed K2 criterion. As a result, the following damage indexes were obtained – 91.43%–97.14% correctly classified patterns for the training set, while for the test set – 79.17%–100%. The conducted research is interdisciplinary in the field of civil engineering and mining and environmental engineering, and the very issue of damage is also important from the socio-economical and energy point of view, especially for a large number of buildings.

Conundrum of fault detection in active hybrid AC–DC distribution networks

Fault detection in hybrid AC–DC distribution networks is a challenging problem due to various sources of uncertainty and high degrees of complexity. A few well-known sources that instil uncertainty in the system are stochasticity of energy injected by distributed energy resources, noisy or corrupt data, heterogeneity of agents, problems with the automated mapping of equipment connectivity, and partial knowledge of the system. This study presents a distinctive approach that draws upon the use of Bayesian belief network to overcome uncertainties. The key advantage of Bayesian inference methodology is its capability to leverage both causal and correlational data in formulating a plausible conclusion. The proposed method uses state variables produced by distributed state estimation along with data collected from self-aware agents as the main sources of causal information. The rationale for using state estimation is its capability to overarch heterogeneity of AC and DC agents. It is shown that probabilistic graphical models can be employed successfully to detect faults in active hybrid distribution networks. An augmented version of IEEE 13-bus network is utilised to simulate and verify the suitability and effectiveness of the proposed technique.

Using Dynamic Bayesian Belief Network for analysing well decommissioning failures and long-term monitoring of decommissioned wells

There is increasing interest to consider dependent failures and human errors in the offshore industry. Permanently abandoned wells dot most of the subsea environment. The nature of a well plugging and abandonment (Well P&A) run is usually conducted in a manner such that the lowest-cost contractor is engaged to plug several wells tapping the same reservoir. Thus, this makes it an ideal case study for incorporating failures based on common causes. The heavy use of operators during a cementing job also provides the case for analysis of human error in such tasks. One proposed method to analyse the above-mentioned is the use of Bayesian Belief Networks (BBN) to achieve the following objectives (1) to capture better estimates of a well PA event by incorporating dependencies, and meet regulatory requirements by authorities; and (2) to use the same model to provide long term monitoring of a group of wells linked by common dependencies. This model has not only captured the dependencies of multiple variables, but also projected it in a dynamic manner to provide a risk profile for the next decade where well integrity failure is likely to happen. The sensitive analysis and backwards diagnostic analysis demonstrated that the results agree with a statistical study of 103 wells. This affirms that well failure probabilities are under-estimated if independent failure is assumed.

Integration of weather conditions for predicting microbial water quality using Bayesian Belief Networks

Levels of fecal indicator bacteria (FIB) provide a surrogate measure of the microbial quality of water used for a wide range of applications. Despite the common use of these measures, a significant limitation is a delay in results due to the time required for cultivation and enumeration of FIB. Testing requires at least 18–24 h, and therefore, FIB cannot be used to identify current or real-time microbial water quality. An approach of nowcasting or empirical modelling approaches that incorporate water quality, environmental, and weather variables to predict FIB levels in real-time has been developed with some success. However, FIB levels are dependent on a complex interaction of numerous variables, which can be challenging to model with ordinary linear regression or classification methods most commonly applied. In this study, novel use of Bayesian Belief Networks (BBNs) that allow for a probabilistic representation of complex variable interactions is investigated for real-time modelling of FIB levels surface waters. In particular, the integration of both water quality measures and current/historical weather for prediction of fecal coliforms and Escherichia coli levels is achieved using BBNs. For 4-bin classification of fecal coliform levels, BBNs increased prediction accuracy by 25%–54% compared to other previously used techniques including logistic regression, Naïve Bayes, and random forests. Binary prediction of E. coli levels exceeding a threshold of 20 CFU/100 mL was also significantly improved using BBNs with prediction accuracies >90% for all monitoring sites. Advantages of the BBN approach are also demonstrated identifying the ability to make predictions from incomplete monitoring data as well as probabilistic inference of variable importance in FIB levels. In particular, the results indicate that water quality surrogates such as conductivity are essential to real-time prediction of FIB. The results and models described in this work can be readily utilized to provide accurate and real-time assessments of FIB levels in surface waters utilizing commonly monitored parameters.

Application of a Bayesian belief network model for assessing the risk of wind erosion: A test with data from wind tunnel experiments

Abstract The complexity of the interactions between drivers in wind erosion processes and the absence of adequate and reliable data are major constraints to achieving a quantitative assessment of wind erosion. Bayesian Belief Networks (BBNs) provide a useful approach to address real-world problems, where available data and knowledge are disparate, limited or uncertain. We investigated the potential use of BBNs to assess soil erosion risk in a typical arid region that experiences severe wind erosion. The developed framework was based on a standard risk assessment procedure, where the risk of wind erosion was quantified by incorporating assessments of consequence and vulnerability. Performance of the constructed model was evaluated using scenario testing, sensitivity analysis, and wind-tunnel measurements. The model provided reasonable estimates of the soil vulnerability, consequence, and risk to/of wind erosion. The results showed that weather and management factors were the most important parameters affecting wind erosion risk. Based on the fitted regression lines, there were positive (R2 = 0.82) and negative (R2 = 0.72) relationships between the measured wind erosion rates and the predicted probabilities to ‘high’ and ‘low’ vulnerability classes, respectively.

Tri-generation investment analysis using Bayesian network: A case study

ABSTRACTThe increasing energy demand, increasing energy dependency, energy supply security, and environmental concerns have become a part of business policies since COP21 agreements in Paris, 2015. Combined cooling, heating, and power (CCHP or tri-generation) systems play an important role in paying the necessary attention to these policies. Tri-generation investment is a complex decision with hybrid use of energy resources. This article aims to reduce the complexity of this decision by the use of Bayesian belief networks in pre-investment stage of tri-generation investment project cycle. The proposed model gives an insight into decision analysis and helps the decision-makers either generate or purchase from it in order to meet the energy demand with different scenarios. The model is studied for a university case. The investment decision for a CCHP (tri-generation) system will be discussed as an alternative for purchasing the electricity and natural gas from the national grids.

Bayesian Belief Network Used in the Chemical and Process Industry: A Review and Application

With the increasing growth of the chemical and process industries, it is necessary to ensure the safe operation of their complex and often hazardous installations, given their proximity to residential areas. Several techniques, such as fault tree analysis (FTA), bow-tie analysis (BTA), and Bayesian belief networks (BBNs), have been developed for adequate probabilistic risk assessment and management. The current work is aimed at performing a brief statistical review of the use of Bayesian networks in the chemical and process industry within the last decade. The review reveals that Bayesian networks have been used extensively in various forms of safety and risk assessment. This trend is attributable to the complexity of the installations found in this industry and the ability of BBN to intuitively represent these complexities, handle uncertainties, and update event probabilities. The paper is concluded with an illustrative example of the use of BBN to investigate the effectiveness of the safety barriers of a gas facility.

A probabilistic data-driven framework for scoring the preoperative recipient-donor heart transplant survival

Recent research has shown that data mining models can accurately predict the outcome of a heart transplant based on predictors that include patient and donor's health/demographics. These models have not been adopted in practice, however, since they did not: a) consider the interactions between the explanatory variables; b) provide a patient's specific risk of survival (reported results have been primarily deterministic); and c) offer an automated decision tool that can provide some data-driven insights to practitioners. In this study, we attempt to overcome these three limitations through the use of Bayesian Belief Networks (BBN). The proposed BBN framework is comprised of four phases. In the first two phases, the data is preprocessed, and a candidate set of predictors is generated based on employing several variable selection methods. The third phase involves the addition of medically relevant variables to the list. In phase four, the BBN model is applied. The results show that the proposed BBN method provides similar predictive performance to the best approaches in the literature. More importantly, our method provides novel information on the interactions among the predictors and the conditional probability of survival for a given set of relevant donor–recipient characteristics. We offer U.S. practitioners a decision support tool that presents an individualized survival score based on our BBN model (and the UNOS dataset).

Developing a Social, Cultural and Economic Report Card for a Regional Industrial Harbour.

Report cards are increasingly used to provide ongoing snap-shots of progress towards specific ecosystem health goals, particularly in coastal regions where planners need to balance competing demands for coastal resources from a range of industries. While most previous report cards focus on the biophysical components of the system, there is a growing interest in including the social and economic implications of ecosystem management to provide a greater social-ecological system understanding. Such a report card was requested on the Gladstone Harbour area in central Queensland, Australia. Gladstone Harbour adjoins the southern Great Barrier Reef, and is also a major industrial and shipping port. Balancing social, economic and environmental interests is therefore of great concern to the regional managers. While environmental benchmarking procedures are well established within Australia (and elsewhere), a method for assessing social and economic performance of coastal management is generally lacking. The key aim of this study was to develop and pilot a system for the development of a report card relating to appropriate cultural, social and economic objectives. The approach developed uses a range of multicriteria decision analysis methods to assess and combine different qualitative and quantitative measures, including the use of Bayesian Belief Networks to combine the different measures and provide an overall quantitative score for each of the key management objectives. The approach developed is readily transferable for purposes of similar assessments in other regions.

Approaches to the Data Coherence Diagnosis in Bayesian Belief Network Models

Bayesian belief networks provide the ability to combine different types of information, e.g. statistical or expert data, allow working with incomplete or inaccurate information; they have clarity and other useful properties. Due to this, Bayesian belief networks have become a popular and highly effective tool in many fields of research. However, in many research areas data provided by the experts can be incoherent, and so in some tasks one should use tools to verify their coherence. The paper discusses examples of application of the Bayesian belief networks in medicine and public health, ecology, economics and risk analysis, functional safety, sociology, and other research areas, and shows the need to develop methods to check the coherence of initial data. The purpose of this work is to systematize problems and examples that illustrate the use of Bayesian belief networks by reviewing and to assess their use of data coherence diagnosis and its importance.

A risk assessment approach to improve the resilience of a seaport system using Bayesian networks

Over the years, many efforts have been focused on developing methods to design seaport systems, yet disruption still occur because of various human, technical and random natural events. Much of the available data to design these systems are highly uncertain and difficult to obtain due to the number of events with vague and imprecise parameters that need to be modelled. A systematic approach that handles both quantitative and qualitative data, as well as means of updating existing information when new knowledge becomes available is required. Resilience, which is the ability of complex systems to recover quickly after severe disruptions, has been recognised as an important characteristic of maritime operations. This paper presents a modelling approach that employs Bayesian belief networks to model various influencing variables in a seaport system. The use of Bayesian belief networks allows the influencing variables to be represented in a hierarchical structure for collaborative design and modelling of the system. Fuzzy Analytical Hierarchy Process (FAHP) is utilised to evaluate the relative influence of each influencing variable. It is envisaged that the proposed methodology could provide safety analysts with a flexible tool to implement strategies that would contribute to the resilience of maritime systems.

A GIS plug-in for Bayesian belief networks: Towards a transparent software framework to assess and visualise uncertainties in ecosystem service mapping

The complexity and spatial heterogeneity of ecosystem processes driving ecosystem service delivery require spatially explicit models that take into account the different parameters affecting those processes. Current attempts to model ecosystem service delivery on a broad, regional scale often depend on indicator-based approaches that are generally not able to fully capture the complexity of ecosystem processes. Moreover, they do not allow quantification of uncertainty on their predictions. In this paper, we discuss a QGIS plug-in which promotes the use of Bayesian belief networks for regional modelling and mapping of ecosystem service delivery and associated uncertainties. Different types of specific Bayesian belief network output maps, delivered by the plug-in, are discussed and their decision support capacities are evaluated. This plug-in, used in combination with firmly developed Bayesian belief networks, has the potential to add value to current spatial ecosystem service accounting methods. The plug-in can also be used in other research domains dealing with spatial data and uncertainty.

IJARCCE - Computer and Communication Engineering

1,2,3,4 Abstract: The project consists of an efficient Currently Many researchers have addressed this important problem with varying end goals and have proposed valuation techniques to compute the total number of faults. A group of researchers focuses on finding error-prone modules base on the size of the module. Briand et al. Predict the number of remaining faults during inspection using actual inspection data, where as Strand et al predict which files will contain the most faults in the next release. Zhang and Mockus use data collected from previous projects to estimate the number of faults in a new project. However, these data sets are not always available or, even if they are, may lead to inaccurate estimates. For example, Zhang and Mockus use a naive method base only on the size of the product to select similar projects while ignoring many other critical factors such as project type, complexity, etc. Another alternative that appears to produce very accurate estimates is base on the use of Bayesian Belief Networks (BBNs) .However, these techniques require the use of additional information, such as expert knowledge and empirical data, that are not necessarily collected by most software development companies. Software reliability growth models (SRGMs) are also used to estimate the total number of faults to measure software reliability. Although they can be used to indicate the status of the testing process, some have slow convergence while others have limited application as they may require more input data or initial values that are selected by experts.

Evaluating acceptability of groundwater protection measures under different agricultural policies

Water resources management is often characterized by conflicts in many arid and semi-arid regions, where agriculture is the main user of groundwater (GW). Conflicts could arise among different decision-makers and stakeholders. Moreover, different policies can interact each other hampering or facilitating their implementation and effectiveness. This contribution describes a new implementation of GeSAP, an integrated modelling tool for enabling local GW management by combining the need for GW protection with socio-economic and behavioural determinants of GW use. GeSAP is based on the involvement of multiple stakeholders and the use of Bayesian Belief Networks (BBN) to simulate and explore their attitude relative to GW exploitation and their responses to the introduction of new protection and agricultural policies. In this work, GeSAP was implemented in the area of the Capitanata Irrigation Users Organization, located in the Apulia region (southern Italy). It was used to simulate the reactions of the main stakeholders involved in GW protection policy implementation and to assess the policy's effectiveness in terms of actual reduction of GW exploitation. Furthermore, the interactions between the GW protection policy and the coming reform of the Common Agricultural Policy (CAP) was investigated. The results of the application proved the capability of the GeSAP tool to assess the actual effectiveness of GW protection policy by investigating how far this policy could be considered acceptable by farmers. In addition, this study demonstrates how the effectiveness of the GW protection policy could be affected by the interaction with the CAP reform. The latter could strongly impact the balance between water demand and availability with the effect of nullifying the positive synergy between CAP and GW protection policy. Although water management issues are not explicitly mentioned among the main scopes of the CAP, this work clearly demonstrates the impact that such policy could have on farmers’ decisions on water use.

REASONING UNDER UNCERTAINTY WITH BAYESIAN BELIEF NETWORKS ENHANCED WITH ROUGH SETS

The objective of this paper is to present a new approach to reasoning under uncertainty, based on the use of Bayesian belief networks (BBN’s) enhanced with rough sets. The role of rough sets is to provide additional reasoning to assist a BBN in the inference process, in cases of missing data or difficulties with assessing the values of related probabilities. The basic concepts of both theories, BBN’s and rough sets, are briefly introduced, with examples showing how they have been traditionally used to reason under uncertainty. Two case studies from the authors’ own research are discussed: one based on the evaluation of software tool quality for use in real-time safety-critical applications, and another based on assisting the decision maker in taking the right course of action, in real time, in the naval military exercise. The use of corresponding public domain software packages based on BBN’s and rough sets is outlined, and their application for real-time reasoning in processes under uncertainty is presented.