Search

This paper presents a decision-making support system for situation risk assessment associated with critical alarms conditions in a gas facility. The system provides a human operator with advice on the confirmation and classification of occurred alarm. The input of the system comprises uncertain and incomplete information. In the light of uncertain and incomplete information, different uncertainties laws have been associated with the probabilistic assessment of the system loops which combine data of several sources to reach the ultimate classification. The implemented model used Observe-Orient-Decide-Act loop (OODA) combined with Bayesian networks. Results show that the system can classify the alarms system.

Expert judgement is a common method for software effort estimations in practice today. Estimators are often shown extra obsolete requirements together with the real ones to be implemented. Only one previous study has been conducted on if such practices bias the estimations. We conducted six experiments with both students and practitioners to study, and quantify, the effects of obsolete requirements on software estimation. By conducting a family of six experiments using both students and practitioners as research subjects ($$N=461$$ N = 461 ), and by using a Bayesian Data Analysis approach, we investigated different aspects of this effect. We also argue for, and show an example of, how we by using a Bayesian approach can be more confident in our results and enable further studies with small sample sizes. We found that the presence of obsolete requirements triggered an overestimation in effort across all experiments. The effect, however, was smaller in a field setting compared to using students as subjects. Still, the over-estimations triggered by the obsolete requirements were systematically around twice the percentage of the included obsolete ones, but with a large 95% credible interval. The results have implications for both research and practice in that the found systematic error should be accounted for in both studies on software estimation and, maybe more importantly, in estimation practices to avoid over-estimations due to this systematic error. We partly explain this error to be stemming from the cognitive bias of anchoring-and-adjustment, i.e. the obsolete requirements anchored a much larger software. However, further studies are needed in order to accurately predict this effect.

A novel multi-criteria decision making approach based on an adaptive copula Bayesian network (CBN) model is proposed to effectively handle complex dependence problems under uncertainty. Specifically, the Bayesian network is used to merge various criteria in a model and graphically describe cause-effect relationships. The copula is incorporated to construct joint distributions of variables by specifying marginal distributions and copula functions separately. Regarding the practical value, the constructed model can adjust to changeable conditions to provide adaptive suggestions in view of diverse disciplines. The effectiveness of the proposed approach is verified in a case study about choosing the most suitable location of the pedestrian overhead bridge (POB) to install lift facilities in Singapore. Firstly, three criteria and ten relevant influential factors concerning the sustainable aspect are derived from a combination of expert knowledge and statistical data. Then, a proper CBN model is developed under the consideration of these determined criteria and factors, aiming to predict the constructability index for alternative locations statistically. Finally, the correlation analysis and CBN inference are performed to evaluate and identify the ideal location in a data-driven manner. Furthermore, results from the proposed CBN-based approach are compared against the traditional Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) to illustrate its advantages in representing dependency, modeling uncertainty, fusing information, and reducing subjectivity. Ministry of Education (MOE) Nanyang Technological University The Ministry of Education Tier 1 Grant, Singapore (No. 04MNP000279C120 and the Start-Up Grant at Nanyang Technological University, Singapore (No. 04INS000423C120) are acknowledged for their financial support of this research.

This paper applies the variational methods to learn the parameters and the probability of evidence of directed graphic models (also known as Bayesian networks (BNs)) when data contains missing values. One class of variational methods, the Bethe/Kikuchi approximate algorithm, is combined with Expectation-Maximization (EM) to learn BN model parameters from data using an upper limit on cluster size. Our proposed method is novel since the Bethe/Kikuchi algorithms are typically only used to approximate marginal distributions. We review popular Bethe/Kikuchi approximate algorithms, including Belief Propagation (BP), Iterative Join Graph Propagation (IJGP), and Fundamental Cycle Base (FCB) algorithms for discrete BNs, and conduct experiments to test their learning performance using a wide range of discrete BNs in practice. Experiments show that by using IJGP or FCB algorithms, with EM, we can obtain more accurate results than the conventional approximate algorithms, such as Evidence Pre-propagation Importance Sampling and Variational Inference. When exact methods are intractable the IJGP/FCB with EM can be an effective alternative solution.

Learning data representations that capture task-related features, but are invariant to nuisance variations remains a key challenge in machine learning. We introduce an automated Bayesian inference framework, called AutoBayes, that explores different graphical models linking classifier, encoder, decoder, estimator and adversarial network blocks to optimize nuisance-invariant machine learning pipelines. Auto Bayes also enables learning disentangled representations, where the latent variable is split into multiple pieces to impose various relationships with the nuisance variation and task labels. We benchmark the framework on several public datasets, and provide analysis of its capability for subject-transfer learning with/without variational modeling and adversarial training. We demonstrate a significant performance improvement with ensemble learning across explored graphical models.

Currently, significant development of methods supporting decision making under uncertainty conditions is observed. One of such methods includes Bayesian networks used in many fields of economy and science. The paper presents the use of the Bayesian network method in civil engineering problems with particular emphasis on construction engineering projects. In addition to the existing examples of the use of the method cited, the authors’ method for the risk estimation of additional works is presented.

The application of data mining techniques and statistical analysis to the sports field has received increasing attention in the last decade. One of the most famous sports in the world is soccer, and the present work deals with it, using data from the 2009/2010 season to the 2015/2016 season from nine European leagues extracted from the Kaggle European Soccer database. Overall performance indicators of the four roles in a soccer team (forward, midfielder, defender and goalkeeper) for home and away teams are used to investigate the relationships between them and the results of matches, and to predict the wins of the home team. The model used to answer both these demands is the Bayesian Network. This study shows that this model can be very useful for mining the relations between players' performance indicators and for improving knowledge of the game strategies applied by coaches in different leagues. Moreover, it is shown that the ability to predict match results of the proposed Bayesian Network is roughly the same as that of the Naive Bayes model.

Information technology is so ubiquitous and AI’s progress so inspiring that also legal professionals experience its benefits and have high expectations. At the same time, the powers of AI have been rising so strongly that it is no longer obvious that AI applications (whether in the law or elsewhere) help promoting a good society; in fact they are sometimes harmful. Hence many argue that safeguards are needed for AI to be trustworthy, social, responsible, humane, ethical. In short: AI should be good for us. But how to establish proper safeguards for AI? One strong answer readily available is: consider the problems and solutions studied in AI & Law. AI & Law has worked on the design of social, explainable, responsible AI aligned with human values for decades already, AI & Law addresses the hardest problems across the breadth of AI (in reasoning, knowledge, learning and language), and AI & Law inspires new solutions (argumentation, schemes and norms, rules and cases, interpretation). It is argued that the study of AI as Law supports the development of an AI that is good for us, making AI & Law more relevant than ever.

Current decision support systems address domains that are heterogeneous in nature and becoming progressively larger. Such systems often require the input of expert judgement about a variety of different fields and an intensive computational power to produce the scores necessary to rank the available policies. Recently, integrating decision support systems have been introduced to enable a formal Bayesian multi-agent decision analysis to be distributed and consequently efficient. In such systems, where different panels of experts independently oversee disjoint but correlated vectors of variables, each expert group needs to deliver only certain summaries of the variables under their jurisdiction, derived from a conditional independence structure common to all panels, to properly derive an overall score for the available policies. Here we present an algebraic approach that makes this methodology feasible for a wide range of modelling contexts and that enables us to identify the summaries needed for such a combination of judgements. We are also able to demonstrate that coherence, in a sense we formalize here, is still guaranteed when panels only share a partial specification of their model with other panel members. We illustrate this algebraic approach by applying it to a specific class of Bayesian networks and demonstrate how we can use it to derive closed form formulae for the computations of the joint moments of variables that determine the score of different policies.

This paper presents a case study demonstrating how to quantify thevalue of structural health monitoring (SHM), when used optimally inmaintenance planning for wind turbine blades. Maintenance costoptimization is performed using a risk-based approach based onBayesian decision analysis, in which probabilistic models are developedfor blade deterioration processes, blade inspections and SHM systems.The probabilistic SHM system model is based on data from an SHMcampaign with a 225 kW Vestas V27 wind turbine, where an artificialtrailing edge crack of increasing size was introduced. The statisticsderived from this model are applied to the case study concerningmaintenance of an 8 MW offshore wind turbine. It is found that thebenefit of the SHM highly depends on the reliability of the SHM systemand on how SHM observations are used when making decisions oninspections and maintenance. A sensitivity study confirms the generalityof the findings. The paper presents a case study demonstrating the methods used to estimate the value of the information (VoI) delivered by a structural health monitoring (SHM) system. The monitored object is a blade of an operating wind turbine. The case study shows how the maintenance cost optimization can be performed using a risk-based approach cast in a Bayesian decision analysis framework, in which probabilistic models are developed for blade deterioration processes, blade inspections, and SHM systems. The paper also provides a parameter study, which demonstrates how the potential benet of SHM highly depends on the reliability of the utilized SHM system and how the SHM observations are used for decision making on the inspections and maintenance.

Funding Information: The work in this paper has been supported by SIMREC project (CBC 2014-2020), funded by the European Union, the Russian Federation and the Republic of Finland. The contributions to this work by the second author were supported by the Canada Research Chairs Program, through a grant from the Natural Sciences and Engineering Research Council (NSERC). The authors also thank the reviewers for their constructive comments which have helped to improve a previous version of this article. Publisher Copyright: © 2021 The Authors Digitalization is becoming a trend in our modern society and systems. Meanwhile, risk analysis and management has rooted and been applied in various fields. Therefore, there is an increasing need to integrate risk analysis and management into the coming digital society. Risk has been represented digitally by the product of probability and consequence i.e. R=P×C traditionally. However, it has been increasingly discussed to include strength of evidence (SoE) in addition to the traditional consequence (C) and probability (P). Although much advance has been achieved along this direction, there still remains challenges, e.g. ambiguity in rating SoE and visual expression of risk diagrams. This paper focuses on addressing these issues and meanwhile aims to make the risk expression fully digital so that it is more efficient and flexible to be included in a system analysis and visualization. This is achieved firstly by reviewing state-of-the-art discussions on SoE assessment in risk management and identifying the remaining challenges. Then, the paper proposes an approach to address the challenges by forming a fuzzy logic SoE index based on fuzzy logic theory, which enables a transfer from linguistic variable to a digital one with the ambiguity avoided. After the SoE index is formed, it is applied into BNs as the node size index to demonstrate its practical application. Meanwhile, with the BNs forming the infrastructure to calculate and present consequences and probabilities, it showcases a new system risk management approach. All the variables in the system can be expressed in a risk diagram. This further enables an improved risk visualization, risk management and risk communication for system analysis, towards risk digitalization.

Background Sustainable energy transition of a country is complex and long-term process, which requires decision-making in all stages and at all levels, including a large number of different factors, with different causality. The main objective of this paper is the development of a probabilistic model for decision-making in sustainable energy transition in developing countries of SE Europe. The model will be developed according to the specificities of the countries for which it is intended—SE Europe. These are countries where energy transition is slower and more difficult due to many factors: high degree of uncertainty, low transparency, corruption, investment problems, insufficiently reliable data, lower level of economic development, high level of corruption and untrained human resources. All these factors are making decision-making more challenging and demanding. Methods Research was done by using content analysis, artificial intelligence methods, software development method and testing. The model was developed by using MSBNx—Microsoft Research’s Bayesian Network Authoring and Evaluation Tool. Results Due to the large number of insufficiently clear, but interdependent factors, the model is developed on the principle of probabilistic (Bayesian) networks of factors of interest. The paper presents the first model for supporting decision-making in the field of energy sustainability for the region of Southeastern Europe, which is based on the application of Bayesian Networks. Conclusion Testing of the developed model showed certain characteristics, discussed in paper. The application of developed model will make it possible to predict the short-term and long-term consequences that may occur during energy transition by varying these factors. Recommendations are given for further development of the model, based on Bayesian networks.

The complicated characteristics of wastewater treatment plants (WWTPs) significantly hinder the monitoring of industrial processes, and thus much attention has been paid to process modeling and prediction. A fuzzy partial least squares-based dynamic Bayesian networks (FPLS-DBN) is proposed to improve the modeling ability in WWTPs. To adapt the nonlinear process data, fuzzy partial least squares (FPLS) is introduced by using a fuzzy system to extract nonlinear features from process data. In addition, a dynamic extension is included by embedding augmented matrices into Bayesian networks to fit the uncertainty and time-varying characteristics. Regarding the quality indices for effluent suspended solid in the WWTP, the root mean square error of the FPLS-DBN model is decreased by 28.63% and 69.47%, respectively, in comparison with that for partial least squares and Bayesian networks. The results demonstrate the superiority of FPLS-DBN in modeling performance for an actual industrial WWTP application.

In the bridge industry, current traffic trends have increased the likelihood of having the simultaneous presence of both extreme live loads and earthquake events. To date, their concurrent interaction has scarcely been systematically studied. Prevailing studies have investigated the isolated existence of either live loads or seismic actions. In an effort to fill this gap in the literature, a non-parametric Bayesian Network (BN) has been proposed. It is aimed at evaluating the conditional probability of failure for a reinforced concrete bridge column, subject simultaneously to the actions mentioned above. Based on actual data from a structure located in the State of Mexico, a Monte Carlo Simulation model was developed. This led to the construction of a BN with 17 variables. The set of variables included in the model can be categorized into three groups: acting loads, materials resistances and structure force-displacement behavior. Practitioners are then provided with a tool for unspecialized labor force to gather information in situ (e.g. Weight-In-Motion data and Schmidt hammer measurements), which can be included in the network, leading to an updated probability of failure. Moreover, this framework also serves as a quantitative tool for bridge column reliability assessments. Results from the theoretical model confirmed that the bridge column probability of failure was within the expected range reported in the literature. This reflects not only the appropriateness of its design but also the suitability of the proposed BN for reliability analysis.

peer-reviewed Inroads have been made in our understanding of the risks posed to human health and the environment by nanoparticles (NPs) but this area requires continuous research and monitoring. Machine learning techniques have been applied to nanotoxicology with very encouraging results. This study deals with bridging physicochemical properties of NPs, experimental exposure conditions and in vitro characteristics with biological effects of NPs on a molecular cellular level from transcriptomics studies. The bridging is done by developing and implementing Bayesian Networks (BNs) with or without data preprocessing. The BN structures are derived either automatically or methodologically and compared. Early stage nanotoxicity measurements represent a challenge, not least when attempting to predict adverse outcomes and modeling is critical to understanding the biological effects of exposure to NPs. The preprocessed data-driven BN showed improved performance over automatically structured BN and the BN with unprocessed datasets. The prestructured BN captures inter relationships between NP properties, exposure condition and in vitro characteristics and links those with cellular effects based on statistic correlation findings. Information gain analysis showed that exposure dose, NP and cell line variables were the most influential attributes in predicting the biological effects. The BN methodology proposed in this study successfully predicts a number of toxicologically relevant cellular disrupted biological processes such as cell cycle and proliferation pathways, cell adhesion and extracellular matrix responses, DNA damage and repair mechanisms etc., with a success rate >80%. The model validation from independent data shows a robust and promising methodology for incorporating transcriptomics outcomes in a hazard and, by extension, risk assessment modeling framework by predicting affected cellular functions from experimental conditions.

© 2019, The Author(s). Machine learning (ML) algorithms steer agent decisions in agent-based models (ABMs), serving as a vehicle for implementing behaviour changes during simulation runs. However, when training an ML algorithm, obtaining large sets of micro-level human behaviour data is often problematic. Information on human behaviour is often collected via surveys of relatively small sample sizes. This paper presents a methodology for training a learning algorithm to guide agent behaviour in a spatial ABM using a limited survey data sample. We apply different implementation strategies using survey data and Bayesian networks (BNs). By being grounded in probabilistic directed graphical models, BNs stand out among other learning algorithms in that they can be based on expert knowledge and/or known datasets. This paper presents four alternative implementations of data-driven BNs to support agent decisions in a spatial ABM. We differentiate between training BNs prior to, or during the simulation runs, using only survey data or a combination of survey data and expert knowledge. The four different implementations are then illustrated using a spatial ABM of cholera diffusion for Kumasi, Ghana. The results indicate that a balance between expert knowledge and survey data provides the best control over the learning process of the agents and produces the most realistic agent behaviour.

Reconstructability Analysis (RA) and Bayesian Networks (BN) are both probabilistic graphical modeling methodologies used in machine learning and artificial intelligence. There are RA models that are statistically equivalent to BN models and there are also models unique to RA and models unique to BN. The primary goal of this paper is to unify these two methodologies via a lattice of structures that offers an expanded set of models to represent complex systems more accurately or more simply. The conceptualization of this lattice also offers a framework for additional innovations beyond what is presented here. Specifically, this paper integrates RA and BN by developing and visualizing: (1) a BN neutral system lattice of general and specific graphs, (2) a joint RA-BN neutral system lattice of general and specific graphs, (3) an augmented RA directed system lattice of prediction graphs, and (4) a BN directed system lattice of prediction graphs. Additionally, it (5) extends RA notation to encompass BN graphs and (6) offers an algorithm to search the joint RA-BN neutral system lattice to find the best representation of system structure from underlying system variables. All lattices shown in this paper are for four variables, but the theory and methodology presented in this paper are general and apply to any number of variables. These methodological innovations are contributions to machine learning and artificial intelligence and more generally to complex systems analysis. The paper also reviews some relevant prior work of others so that the innovations offered here can be understood in a self-contained way within the context of this paper.

We exploit the use of a controller area network (CAN-bus) to monitor sensors on the buses of local public transportation in a big European city. The aim is to advise fleet managers and policymakers on how to reduce fuel consumption so that air pollution is controlled and public services are improved. We deploy heuristic algorithms and exhaustive ones to generate Bayesian networks among the monitored variables. The aim is to describe the relevant relationships between the variables, to discover and confirm the possible cause–effect relationships, to predict the fuel consumption dependent on the contextual conditions of traffic, and to enable an intervention analysis to be conducted on the variables so that our goals are achieved. We propose a validation technique using Bayesian networks based on Granger causality: it relies upon observations of the time series formed by successive values of the variables in time. We use the same method based on Granger causality to rank the Bayesian networks obtained as well. A comparison of the Bayesian networks discovered against the ground truth is proposed in a synthetic data set, specifically generated for this study: the results confirm the validity of the Bayesian networks that agree on most of the existing relationships.

Since recording the first case in Wuhan in November 2020, COVID-19 is still spreading widely and rapidly affecting the health of millions all over the globe. For fighting against this pandemic, numerous strategies have been made, where the early isolation is considered among the most effective ones. Proposing useful methods to screen and diagnose the patient’s situation for the purpose of specifying the adequate clinical management represents a significant challenge in diminishing the rates of mortality. Inspired from this current global health situation, we introduce a new autonomous process of decision-making that consists of two modules. The first module is the data analysis based on Bayesian network that is employed to indicate the coronavirus symptoms severity and then classify COVID-19 cases as severe, moderate or mild. The second module represents the decision-making based on association rules method that generates autonomously the adequate decision. To construct the model of Bayesian network, we used an effective method-oriented data for the sake of learning its structure. As a result, the algorithm accuracy in making the correct decision is 30% and in making the adequate decision is 70%. These experimental results demonstrate the importance of the suggested methods for decision-making.

For an AI agent to make trustworthy decision recommendations under uncertainty on behalf of human principals, it should be able to explain why its recommended decisions make preferred outcomes more likely and what risks they entail. Such rationales use causal models to link potential courses of action to resulting outcome probabilities. They reflect an understanding of possible actions, preferred outcomes, the effects of action on outcome probabilities, and acceptable risks and trade-offs—the standard ingredients of normative theories of decision-making under uncertainty, such as expected utility theory. Competent AI advisory systems should also notice changes that might affect a user’s plans and goals. In response, they should apply both learned patterns for quick response (analogous to fast, intuitive “System 1” decision-making in human psychology) and also slower causal inference and simulation, decision optimization, and planning algorithms (analogous to deliberative “System 2” decision-making in human psychology) to decide how best to respond to changing conditions. Concepts of conditional independence, conditional probability tables (CPTs) or models, causality, heuristic search for optimal plans, uncertainty reduction, and value of information (VoI) provide a rich, principled framework for recognizing and responding to relevant changes and features of decision problems via both learned and calculated responses. This paper reviews how these and related concepts can be used to identify probabilistic causal dependencies among variables, detect changes that matter for achieving goals, represent them efficiently to support responses on multiple time scales, and evaluate and update causal models and plans in light of new data. The resulting causally explainable decisions make efficient use of available information to achieve goals in uncertain environments.

Approximate Computing (AxC) trades off between the accuracy required by the user and the precision provided by the computing system to achieve several optimizations such as performance improvement, energy, and area reduction. Several AxC techniques have been proposed so far in the literature. They work at different abstraction levels and propose both hardware and software implementations. The standard issue of all existing approaches is the lack of a methodology to estimate the impact of a given AxC technique on the application-level accuracy. This paper proposes a probabilistic approach based on Bayesian networks to quickly estimate the impact of a given approximation technique on application-level accuracy. Moreover, we have also shown how Bayesian networks allow a backtrack analysis that automatically identifies the most sensitive components. That influence analysis dramatically reduces the space exploration for approximation techniques. Preliminary results on a simple artificial neural network shown the efficiency of the proposed approach.

In complex systems with stochastically dependent components which are not observed directly, determining an effective maintenance policy is a difficult task. In this paper, a dynamic Bayesian network based maintenance decision framework is proposed to evaluate proactive maintenance policies for such systems. Two preventive and one predictive maintenance strategies from a cost perspective are designed for multi-component dependable systems which aim to reduce maintenance cost while increasing system reliability at the same time. Tabu procedure is employed to avoid repetitive similar actions. The performances of the policies are compared with a reactive maintenance strategy and also with each other using different strategy parameters on a real life system confronted in thermal power plants for six different scenarios. The scenarios are designed considering different structures of system dependability and reactive cost. The results show that the threshold based maintenance which is the predictive strategy gives the minimum cost and maintenance number in almost all scenarios. This research is supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under grant: 117M587. Publisher's Version

During the evolutionary process, algorithms based on probability distributions for generating new individuals suffer from computational burden due to the intensive computation of probability distribution estimations, particularly when using Probabilistic Graph Models (PGMs). In the Bayesian Optimisation Algorithm (BOA), for instance, determining the optimal Bayesian network structure by a given solution sample is an NP-hard problem. To overcome this issue, we consider a new BOA-based optimisation approach (FBOA) which explores the fact that patterns of PGM adjustments can be used as a guide to reduce the frequency of PGM updates because significant changes in PGM structure might not occur so frequently, and because they can be particularly sparse at the end of evolution. In the present paper, this new approach is scrutinised in the search space of an NK-landscape optimisation problem for medium and large-size instances. Average gaps and success rates as well as the correlation between the landscape ruggedness of the problem and the expected runtime of FBOA and BOA are presented for medium-size instances. For large-size instances, optimisation results from FBOA and BOA are compared. The experiments show that, despite our FBOA being of almost three times faster than BOA, it still produces competitive optimisation results.

Freshwater ecosystems are negatively affected by climate change and human interventions modifying together supply and demand of ecosystem services (ES). Research on ES focused on assessing risks arising from the interaction among both stressors, integrating empirical data with expert knowledge. This work aims at incorporating Bayesian Networks (BN) approaches into ES appraisal, identifying key factors driving changes and trade-offs among ES potential under different scenarios. Applying the designed BN to the Taro River basin (TRB) in Italy, the outcomes showed a limited space to improve ES potential, as well as trade-offs between water yield and nutrient retention services due to changes in precipitation and land use patterns. Moreover, the analysis of key input variables highlighted that precipitation is the main driver affecting provisioning services while land use for the regulating ones. The results imply a low capacity to provide services in the medium term for the TRB where water was exploited for multiple competing objectives. Therefore, “win-win” spatial planning and water management strategies are needed to improve freshwater ES potential. The designed BN model represents a valuable decision support tool to quickly perform ES assessment and to identify the most suitable management plan to maintain benefits from freshwater ecosystems.

Air traffic networks, where airports are the nodes that interconnect the entire system, have a time-varying and stochastic nature. An incident in the airport environment may easily propagate through the network and generate system-level effects. This paper analyses the aircraft flow through the Airport Transit View framework, focusing on the airspace/airside integrated operations. In this analysis, we use a dynamic spatial boundary associated with the Extended Terminal Manoeuvring Area concept. Aircraft operations are characterised by different temporal milestones, which arise from the combination of a Business Process Model for the aircraft flow and the Airport Collaborative Decision-Making methodology. Relationships between factors influencing aircraft processes are evaluated to create a probabilistic graphical model, using a Bayesian network approach. This model manages uncertainty and increases predictability, hence improving the system&rsquo, s robustness. The methodology is validated through a case study at the Adolfo Su&aacute, rez Madrid-Barajas Airport, through the collection of nearly 34,000 turnaround operations. We present several lessons learned regarding delay propagation, time saturation, uncertainty precursors and system recovery. The contribution of the paper is two-fold: it presents a novel methodological approach for tackling uncertainty when linking inbound and outbound flights and it also provides insight on the interdependencies among factors driving performance.

Natural and human-made disasters can disrupt infrastructures even if they are designed to be hazard resistant. While the occurrence of hazards can only be predicted to some extent, their impact can be managed by increasing the emergency response and reducing the vulnerability of infrastructure. In the context of risk management, the ability of infrastructure to withstand damage and re-establish their initial condition has recently gained prominence. Several resilience strategies have been investigated by numerous scholars to reduce disaster risk and evaluate the recovery time following disastrous events. A key parameter to quantify the seismic resilience of infrastructures is the Downtime (DT). Generally, DT assessment is challenging due to the parameters involved in the process. Such parameters are highly uncertain and therefore cannot be treated in a deterministic manner. This paper proposes a Bayesian Network (BN) probabilistic approach to evaluate the DT of selected infrastructure types following earthquakes. To demonstrate the applicability of the methodology, three scenarios are performed. Results show that the methodology is capable of providing good estimates of infrastructure DT despite the uncertainty of the parameters. The methodology can be used to effectively support decision-makers in managing and minimizing the impacts of earthquakes in immediate post-event applications as well as to promptly recover damaged infrastructure.

Learning the structure of dependencies among multiple random variables is a problem of considerable theoretical and practical interest. Within the context of Bayesian Networks, a practical and surprisingly successful solution to this learning problem is achieved by adopting score-functions optimisation schema, augmented with multiple restarts to avoid local optima. Yet, the conditions under which such strategies work well are poorly understood, and there are also some intrinsic limitations to learning the directionality of the interaction among the variables. Following an early intuition of Friedman and Koller, we propose to decouple the learning problem into two steps: first, we identify a partial ordering among input variables which constrains the structural learning problem, and then propose an effective bootstrap-based algorithm to simulate augmented data sets, and select the most important dependencies among the variables. By using several synthetic data sets, we show that our algorithm yields better recovery performance than the state of the art, increasing the chances of identifying a globally-optimal solution to the learning problem, and solving also well-known identifiability issues that affect the standard approach. We use our new algorithm to infer statistical dependencies between cancer driver somatic mutations detected by high-throughput genome sequencing data of multiple colorectal cancer patients. In this way, we also show how the proposed methods can shade new insights about cancer initiation, and progression. Code: https://github.com/caravagn/Bootstrap-based-Learning

International audience; We consider the interest of leveraging information between related tasks for learning Bayesian network structures. We propose a new algorithm called Multi-Task Max-Min Hill Climbing (MT-MMHC) that combines ideas from transfer learning, multi-task learning, constraintbased and search-and-score techniques. This approach consists in two main phases. The first one identifies the most similar tasks and uses their similarity to learn their corresponding undirected graphs. The second one directs the edges with a Greedy Search combined with a Branch-and-Bound algorithm. Empirical evaluation shows that MT-MMHC can yield better results than learning the structures individually or than the stateof-the-Art MT-GS algorithm in terms of structure learning accuracy and computational time.

The goal of statistical matching, at a macro level, is the estimation of the joint distribution of variables separately observed in independent samples. The lack of joint information on the variables of interest leads to uncertainty about the data generating model. In this paper we propose the use of graphical models to deal with the statistical matching uncertainty for multivariate categorical variables. The use of Bayesian networks in the statistical matching context allows both to introduce extra sample information on the dependence structure between the variables of interest and to use such an information to factorize the joint probability distribution according to the graph decomposition of a multivariate dependence in lower dimension components. This representation of the joint probability distribution, taking advantage of local relationships, allows to simplify both parameters estimation and statistical matching quality evaluation in a multivariate context. A simulation experiment is performed in order to evaluate the performance of the proposed methodology with and without auxiliary information, as well as to compare it with the saturated multinomial model, in terms of uncertainty reduction. Finally, an application to a real case is provided. Results show a considerable improvement in the quality of statistical matching when the dependence structure is taken into account.

Fast replacement models (or surrogates) have been widely applied in the recent years to accelerate simulation-driven design procedures in microwave engineering. The fundamental reason is a considerable-and often prohibitive-CPU cost of massive full-wave electromagnetic (EM) analyses related to solving common tasks such as parametric optimization or uncertainty quantification. The most popular class of surrogates are data-driven models, which are fast to evaluate, versatile, and easy to handle. Notwithstanding, the curse of dimensionality as well as the utility demands (e.g., so that the model covers sufficiently broad ranges of the system operating conditions), limit the applicability of conventional methods. A performance-driven modeling paradigm allows for mitigating these issue by focusing the surrogate setup process in a constrained domain encapsulating designs being of high quality w.r.t. the assumed figures of interest. The nested kriging framework capitalizing on this idea, renders the constrained surrogate using kriging interpolation, and has been shown to surpass traditional approaches. In pursuit of further accuracy improvements, this work incorporates the performance-driven concept into the fully-connected regression model (FRCM). The latter has been recently introduced in the context of frequency selective surfaces, and combined deep neural networks with Bayesian optimization, the latter employed to determine the network architecture and hyper-parameters. Using two examples of miniaturized microstrip couplers, our methodology is demonstrated to outperform both conventional modeling techniques and nested kriging, with reliable models constructed over multi-dimensional parameters spaces using just a few hundreds of samples. Icelandic Centre for Research (RANNIS) [206606051]; National Science Centre of PolandNational Science Centre, Poland [2020/37/B/ST7/01448] This work was supported in part by the Icelandic Centre for Research (RANNIS) under Grant 206606051, and in part by the National Science Centre of Poland under Grant 2020/37/B/ST7/01448.

Multi-class classification in imbalanced datasets is a challenging problem. In these cases, common validation metrics (such as accuracy or recall) are often not suitable. In many of these problems, often real-world problems related to health, some classification errors may be tolerated, whereas others are to be avoided completely. Therefore, a cost-sensitive variable selection procedure for building a Bayesian network classifier is proposed. In it, a flexible validation metric (cost/loss function) encoding the impact of the different classification errors is employed. Thus, the model is learned to optimize the a priori specified cost function. The proposed approach was applied to forecasting an air quality index using current levels of air pollutants and climatic variables from a highly imbalanced dataset. For this problem, the method yielded better results than other standard validation metrics in the less frequent class states. The possibility of fine-tuning the objective validation function can improve the prediction quality in imbalanced data or when asymmetric misclassification costs have to be considered.

Modern deep learning models achieve state-of-the-art results for many tasks in computer vision, such as image classification and segmentation. However, its adoption into high-risk applications, e.g. automated medical diagnosis systems, happens at a slow pace. One of the main reasons for this is that regular neural networks do not capture uncertainty. To assess uncertainty in classification, several techniques have been proposed casting neural network approaches in a Bayesian setting. Amongst these techniques, Monte Carlo dropout is by far the most popular. This particular technique estimates the moments of the output distribution through sampling with different dropout masks. The output uncertainty of a neural network is then approximated as the sample variance. In this paper, we highlight the limitations of such a variance-based uncertainty metric and propose an novel approach. Our approach is based on the overlap between output distributions of different classes. We show that our technique leads to a better approximation of the inter-class output confusion. We illustrate the advantages of our method using benchmark datasets. In addition, we apply our metric to skin lesion classification—a real-world use case—and show that this yields promising results.

Assistance systems for people with Cerebrovascular Accident (CVA) after-effects such as dysarthria are gaining interest due to the increasing proportion of the population with these disorders. This paper proposes the acquisition and processing of surface Electromyography (sEMG) on the facial muscles activated during the process of speaking and Electroencephalography (EEG), synchronized in time by means of an audio file. Data was collected in seven healthy volunteers and seven patients with dysarthria at the Physiotherapy Department of the Porto Alegre Clinical Hospital. The main objective is to classify the biosignals for verbal commands established by Support Vector Machine (SVM) method for the sEMG and Naive Bayes (NB) method for the EEG. Both methods were compared with the Linear Discriminant Analysis (LDA) method. Extracted features of sEMG signal were: standard deviation, arithmetic mean, skewness, kurtosis and RMS. For the EEG signal, characteristics were extracted in frequency domain such as: minimum, maximum, average and standard deviation, and skewness and kurtosis, in the time domain. As part of the processing method, Common Spatial Pattern (CSP) filters were employed in order to increase the discriminating activity between motion classes in the EEG signal. Mental task identification in healthy and dysarthric subjects reached an accuracy of 77.6–80.8%.

In this work, we investigate an Information Fusion architecture based on a Factor Graph in Reduced Normal Form. This paradigm permits to describe the fusion in a completely probabilistic framework and the information related to the different features are represented as messages that flow in a probabilistic network. In this way we build a sort of context for observed features conferring to the solution a great flexibility for managing different type of features with wrong and missing values as required by many real applications. Moreover, modifying opportunely the messages that flow into the network, we obtain an effective way to condition the inference based on the different reliability of each information source or in presence of single unreliable signal. The proposed architecture has been used to fuse different detectors for an identity document classification task but its flexibility, extendibility and robustness make it suitable to many real scenarios where the signal can be wrongly received or completely missing.

The rivers of KwaZulu-Natal, South Africa, are being impacted by various anthropogenic activities that threaten their sustainability. Our study demonstrated how Bayesian networks could be used to conduct an environmental risk assessment of macroinvertebrate biodiversity and their associated ecosystem to assess the overall effects of these anthropogenic stressors in the rivers. We examined the exposure pathways through various habitats in the study area using a conceptual model that linked the sources of stressors through cause-effect pathways. A Bayesian network was constructed to represent the observed complex interactions and overall risk from water quality, flow and habitat stressors. The model outputs and sensitivity analysis showed ecosystem threat and river health (represented by macroinvertebrate assessment index – MIRAI) could have high ecological risks on macroinvertebrate biodiversity and the ecosystem, respectively. The results of our study demonstrated that Bayesian networks can be used to calculate risk for multiple stressors and that they are a powerful tool for informing future strategies for achieving best management practices and policymaking. Apart from the current scenario, which was developed from field data, we also simulated three other scenarios to predict potential risks to our selected endpoints. We further simulated the low and high risks to the endpoints to demonstrate that the Bayesian network can be an effective adaptive management tool for decision making.

Improved health care services can benefit from a more seamless exchange of medical information between patients and health care providers. This exchange is especially important considering the increasing trends in mobility, comorbidity and outbreaks. However, current Electronic Health Records (EHR) tend to be institution-centric, often leaving the medical information of the patient fragmented and more importantly inaccessible to the patient for sharing with other health providers in a timely manner. Nearly a decade ago, several client&ndash, server models for personal health records (PHR) were proposed. The aim of these previous PHRs was to address data fragmentation issues. However, these models were not widely adopted by patients. This paper discusses the need for a new PHR model that can enhance the patient experience by making medical services more accessible. The aims of the proposed model are to (1) help patients maintain a complete lifelong health record, (2) facilitate timely communication and data sharing with health care providers from multiple institutions and (3) promote integration with advanced third-party services (e.g., risk prediction for chronic diseases) that require access to the patient&rsquo, s health data. The proposed model is based on a Peer-to-Peer (P2P) network as opposed to the client&ndash, server architecture of the previous PHR models. This architecture consists of a central index server that manages the network and acts as a mediator, a peer client for patients and providers that allows them to manage health records and connect to the network, and a service client that enables third-party providers to offer services to the patients. This distributed architecture is essential since it promotes ownership of the health record by the patient instead of the health care institution. Moreover, it allows the patient to subscribe to an extended range of personalized e-health services.

International audience; Through the diversity of criteria and stakes, the uncertain nature of the entailed phenomena and the multi-scale aspects to be taken into account, a river restoration project can be considered as a complex problem. Integrative approaches and modelling tools are thus needed to help river managers make predictions on the evolution of hydromorphological, socio-economic, safety and ecological issues. Such approach can provide valuable information for handling long-term management plans that consider the interaction and the balance of stakeholders interests and river system functioning. In this paper, we present a probabilistic participatory modelling (PM) method that assesses the effects of given restoration actions, knowing the hydromorphological modifications that they may induce on the safety, ecological and socio-economic aspects with the help of local stakeholders through several workshops. To support this strategy, we used Bayesian networks (BNs) as modelling tools as their causal graphs can combine multidimensional knowledge and data from diverse natures. We introduce the causal graphs elaborated with the help of the stakeholders and convert it into BNs that can assist restoration decisions by considering the available decision and utility functions to provide guidance to decision-makers. This was applied to the “Lac des Gaves” reach in the Hautes-Pyrénées, France, a reach that has gone through severe sediment extractions for over 50 years. Each network represents possible restoration decisions linked to one of the observed issues. The paper demonstrates how BNs used as a decision support system (DSS) can help to assess the influence of given management strategies on the river system with the consideration of stakeholders’ knowledge and integration in all the modelling process.

International audience; The project management field has the imperative to increase the success probability of projects. Experts have developed several Project Management Maturity (PMM) models to assess project management practices and improve the project outcome. However, the current literature lacks models that allow experts to correlate the measured maturity with the expected probability of success. The present paper develops a general framework and a method to estimate the impact of PMM on project performance. It uses Bayesian networks to formalize project management experts' knowledge and to extract knowledge from a database of past projects. An industrial case concerning large projects in the oil and gas industry is used to illustrate the application of the method to reduce the risk of project cost (or budget) overruns.

We present a novel bi-objective approach to address the data-driven learning problem of Bayesian networks. Both the log-likelihood and the complexity of each candidate Bayesian network are considered as objectives to be optimized by our proposed algorithm named Nondominated Sorting Genetic Algorithm for learning Bayesian networks (NS2BN) which is based on the well-known NSGA-II algorithm. The core idea is to reduce the implicit selection bias-variance decomposition while identifying a set of competitive models using both objectives. Numerical results suggest that, in stark contrast to the single-objective approach, our bi-objective approach is useful to find competitive Bayesian networks especially in the complexity. Furthermore, our approach presents the end user with a set of solutions by showing different Bayesian network and their respective MDL and classification accuracy results.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. This research proposes a prediction model that might help reducing the dropout rate of university students in Peru. For this, a three-phase predictive analysis model was designed which was combined with the stages proposed by the IBM SPSS Modeler methodology. Bayesian network techniques was compared with decision trees for their level of accuracy over other algorithms in an Educational Data Mining (EDM) scenario. Data were collected from 500 undergraduate students from a private university in Lima. The results indicate that Bayesian networks behave better than decision trees based on metrics of precision, accuracy, specificity, and error rate. Particularly, the accuracy of Bayesian networks reaches 67.10% while the accuracy for decision trees is 61.92% in the training sample for iteration with 8:2 rate. On the other hand, the variables athletic person (0.30%), own house (0.21%), and high school grades (0.13%) are the ones that contribute most to the prediction model for both Bayesian networks and decision trees.

Personal assistive robots to be realized in the near future should have the ability to seamlessly coexist with humans in unconstrained environments, with the robot’s capability to understand and interpret the human behavior during human–robot cohabitation significantly contributing towards this end. Still, the understanding of human behavior through a robot is a challenging task as it necessitates a comprehensive representation of the high-level structure of the human’s behavior from the robot’s low-level sensory input. The paper at hand tackles this problem by demonstrating a robotic agent capable of apprehending human daily activities through a method, the Interaction Unit analysis, that enables activities’ decomposition into a sequence of units, each one associated with a behavioral factor. The modelling of human behavior is addressed with a Dynamic Bayesian Network that operates on top of the Interaction Unit, offering quantification of the behavioral factors and the formulation of the human’s behavioral model. In addition, light-weight human action and object manipulation monitoring strategies have been developed, based on RGB-D and laser sensors, tailored for onboard robot operation. As a proof of concept, we used our robot to evaluate the ability of the method to differentiate among the examined human activities, as well as to assess the capability of behavior modeling of people with Mild Cognitive Impairment. Moreover, we deployed our robot in 12 real house environments with real users, showcasing the behavior understanding ability of our method in unconstrained realistic environments. The evaluation process revealed promising performance and demonstrated that human behavior can be automatically modeled through Interaction Unit analysis, directly from robotic agents.

Prolonging the network lifetime is one of the vital challenges in wireless sensor networks (WSNs). Typically, the lifespan of WSNs can be increased by a technique called clustering, which plays a significant role in simplifying intra-domain routing. The clustering method accounts only a small number of nodes, which are randomly selected as cluster heads (CHs). The main responsibility of CHs is to receive collected data or information from its member nodes and to aggregate the received data and convey the received data to the sink (Sk) or base station (BSn). In this paper, we have proposed a method namely “reliability-based enhanced technique for the ordering of preference by similarity ideal solution (RE-TOPSIS)” combining with fuzzy logic which uses multi-criteria decision making (MCDM) approach aiding in the effective and reliable selection of CHs. It also uses the conventional LEACH protocol to enable one-time CH selection or scheduling in each cluster based on RE-TOPSIS rank index value. This process completely eliminates the need of CH selection process in each round of LEACH's setup state cycle. We have accounted for various criteria such as 1) residual energy; 2) distances between adjacent nodes; 3) energy utilization rates; 4) availability of neighboring nodes; 5) distances between the sink and CHs as well as distances between CHs to member nodes; and 6) the reliability index for completely devising the new scheme. The simulations are accomplished to assess or suggest the performances of the proposed RE-TOPSIS and to compare its performances with the performances of the existing protocols. The results show that the proposed scheme enhances the network lifespan, conserves energy, and introduces a considerable reduction in the frequency of CH selection per cycle by about 20%-25% as compared with the contemporary fuzzy-TOPSIS and LEACH protocols and finally the metrics of the proposed RE-TOPSIS are highlighted.

In China, high-speed rail projects have brought huge social and economic benefits to the affected regions after they are completed. However, the potential externalities of such projects cause competition for the station during the project planning phase, thus triggering social risks. This paper studies the mechanisms responsible for generating the social risk associated with such high-speed rail projects. Employing typical case studies, a social risk list for a given project is established. Based on the risk list, a Bayesian network model is developed and verified through case studies, expert interviews, and expert grading. Using the model&rsquo, s functions of reverse inference and sensitivity analysis, the key risk factors, sensitive risk factors, and maximum causal chain are identified. Countermeasures are then proposed to mitigate the social risk, such as increasing the transparency of and democratizing the planning process for high-speed rail projects, improving the mechanism by which local governments can express interest in such projects, and enhancing emergency management mechanisms. The findings provide points of reference for social risk management when it comes to planning high-speed rail projects and, more generally, offer significant guidance for socially sustainable decision-making processes for mega projects with massive externalities.

The zeta and Moebius transforms over the subset lattice of $n$ elements and the so-called subset convolution are examples of unary and binary operations on set functions. While their direct computation requires $O(3^n)$ arithmetic operations, less naive algorithms only use $2^n \mathrm{poly}(n)$ operations, nearly linear in the input size. Here, we investigate a related $n$-ary operation that takes $n$ set functions as input and maps them to a new set function. This operation, we call multi-subset transform, is the core ingredient in the known inclusion--exclusion recurrence for weighted sums over acyclic digraphs, which extends Robinson's recurrence for the number of labelled acyclic digraphs. Prior to this work the best known complexity bound was the direct $O(3^n)$. By reducing the task to multiple instances of rectangular matrix multiplication, we improve the complexity to $O(2.985^n)$., 12 pages, 4 figures. Appeared in Scandinavian Symposium and Workshops on Algorithm Theory 2020. The replacement is based on the conference version, and has small changes compared to the previous version, particularly to the proof of proposition 13

Conditioned-based maintenance and prognostics and health management enable to optimize maintenance by scheduling the necessary repairs and replacements of technical system components according to their present and future health states. The assessment of future health states is the prognostics and health management keystone. Many technical production systems are made of numerous components implementing their functions. A method to assess the ability of multicomponent systems to carry out future production tasks is proposed to provide decision supports for production and maintenance planning for a better compromise between their objectives. It is based on components prognoses. To handle inherent uncertainties of these prognoses, the method is based on the Dempster Shafer theory and Bayesian networks inferences. Local prognoses are categorized and transformed to be compliant to Dempster Shafer theory. Patterns of systems are identified for which inferences are defined. The patterns are then used to model systems and to assess their abilities to achieve future tasks. An identification of components that should first undergo maintenance is proposed. An example implementing a fictitious complex systems is presented to show how the provided decision supports can be used for production and maintenance planning purposes.

The modeling of phased-mission systems is difficult and the solving process is complex because of the relevance of the phase tasks and the sharing of components existing in different phases or between phases. To solve the problem, based on the cumulative exposure model, the path sets combination method of phased-mission systems is proposed. Aiming at the problem of the cross-stage correlation of components and its different failure rate in each phase, the cumulative exposure model considering the historical damage of components is used to solve by obtaining the cumulative damage distribution of each component in each phase. Firstly, a phased-mission systems reliability model is build by mapping phased-mission system fault trees into a Bayesian network. By traversing the Bayesian network, the minimal path sets of each phase are obtained. Secondly, the disjoint formulas introduced by variable elimination method are used to do the disjoint operation of the minimal path sets of each phase and the conditional probability relations of the common components are used to reduce the minimal path sets scale. Finally, the minimum disjoint path sets of each phase are combined and summed according to the component conditional probability relation. The path sets combination method of phased-mission systems avoids the large conditional probability table, large storage and large computation problems caused by the excessive discrete states in the traditional Bayesian method and the problem that the PMS-BDD method has strict requirements for variable ordering and is difficult to solve the system reliability with multiple failure distribution types of components. In the end, a phased-mission systems reliability modeling and solving is carried out for a geosynchronous orbit satellite, and compared with the PMS-BDD method, which verifies the correctness of the method.

In this paper we propose a scalable importance sampling algorithm for computing Gaussian mixture posteriors in conditional linear Gaussian Bayesian networks. Our contribution is based on using a stochastic gradient ascent procedure taking as input a stream of importance sampling weights, so that a mixture of Gaussians is dynamically updated with no need to store the full sample. The algorithm has been designed following a Map/Reduce approach and is therefore scalable with respect to computing resources. The implementation of the proposed algorithm is available as part of the AMIDST open-source toolbox for scalable probabilistic machine learning ( http://www.amidsttoolbox.com ).

The detection and subsequent reconstruction of incongruent data in time series by means of observation of statistically related information is a recurrent issue in data validation. Unlike outliers, incongruent observations are not necessarily confined to the extremes of the data distribution. Instead, these rogue observations are unlikely values in the light of statistically related information. This paper proposes a multiresolution Bayesian network model for the detection of rogue values and posterior reconstruction of the erroneous sample for non-stationary time-series. Our method builds local Bayesian Network models that best fit to segments of data in order to achieve a finer discretization and hence improve data reconstruction. Our local multiscale approach is compared against its single-scale global predecessor (assumed as our gold standard) in the predictive power and of this, both error detection capabilities and error reconstruction capabilities are assessed. This parameterization and verification of the model are evaluated over three synthetic data source topologies. The virtues of the algorithm are then further tested in real data from the steel industry where the aforementioned problem characteristics are met but for which the ground truth is unknown. The proposed local multiscale approach was found to dealt better with increasing complexities in data topologies.

© 2018 The Author(s). Background: Millions of people worldwide are exposed to deadly infectious diseases on a regular basis. Breaking news of the Zika outbreak for instance, made it to the main media titles internationally. Perceiving disease risks motivate people to adapt their behavior toward a safer and more protective lifestyle. Computational science is instrumental in exploring patterns of disease spread emerging from many individual decisions and interactions among agents and their environment by means of agent-based models. Yet, current disease models rarely consider simulating dynamics in risk perception and its impact on the adaptive protective behavior. Social sciences offer insights into individual risk perception and corresponding protective actions, while machine learning provides algorithms and methods to capture these learning processes. This article presents an innovative approach to extend agent-based disease models by capturing behavioral aspects of decision-making in a risky context using machine learning techniques. We illustrate it with a case of cholera in Kumasi, Ghana, accounting for spatial and social risk factors that affect intelligent behavior and corresponding disease incidents. The results of computational experiments comparing intelligent with zero-intelligent representations of agents in a spatial disease agent-based model are discussed. Methods: We present a spatial disease agent-based model (ABM) with agents' behavior grounded in Protection Motivation Theory. Spatial and temporal patterns of disease diffusion among zero-intelligent agents are compared to those produced by a population of intelligent agents. Two Bayesian Networks (BNs) designed and coded using R and are further integrated with the NetLogo-based Cholera ABM. The first is a one-tier BN1 (only risk perception), the second is a two-tier BN2 (risk and coping behavior). Results: We run three experiments (zero-intelligent agents, BN1 intelligence and BN2 intelligence) and report the results per experiment in terms of several macro metrics of interest: an epidemic curve, a risk perception curve, and a distribution of different types of coping strategies over time. Conclusions: Our results emphasize the importance of integrating behavioral aspects of decision making under risk into spatial disease ABMs using machine learning algorithms. This is especially relevant when studying cumulative impacts of behavioral changes and possible intervention strategies.

[EN] Nowadays, sustainability assessment tends to focus on the biophysical and economic aspects of the built environment. The social aspects are generally overestimated during an infrastructure evaluation. This study proposes a method to optimize infrastructure projects by assessing their social contribution. This proposal takes into account the infrastructure¿s interactions with the local environment in terms of its potential contribution in the short and long term. The method is structured in three stages: (1) preparation of a decision-making model, (2) formulation of the model, and (3) implementation of the model through optimization of infrastructure projects from the social sustainability viewpoint. The theory of Bayesian reasoning and a harmony search optimization algorithm are used to carry out the research. The paper presents the application to a case study of a set of alternatives for road infrastructure projects in El Salvador. This approach creates a model of participative decision-making. The results show that the method can distinguish socially efficient alternatives from the short and long-term contributions. In addition, the results suggest that some variables are less sensitive to the short and long-term maximization, while others vary their values to improve one objective or the other. The findings are directly applied to a real case. The method can be employed in the infrastructure formulation and prioritization phases and complemented with economic and environmental sustainability assessments., This research was funded by the Government of Chile under the Doctoral Fellowship Abroad Program (grant CONICYT-2015/72160059) and the Spanish Ministry of Economy and Competitiveness along with FEDER funding (project BIA2014-56574-R). The authors are grateful to both the United Nations Development Programme and the Ministry of Public Works, Transportation, Housing and Urban Development of El Salvador for their participation in the implementation of the proposed method (Project 00074250).