Decision support under uncertainties based on robust Bayesian networks in reverse logistics management

We elicit numerical expectations for late-onset dementia in the Health and Retirement Study. Our elicitation distinguishes between precise and imprecise probabilities, while accounting for rounding of reports. Respondents quantify imprecision using probability intervals. Nearly half of respondents hold imprecise dementia probabilities, while almost a third of precise-probability respondents round their reports. We provide the first empirical evidence on dementia-risk perceptions among dementia-free older Americans and novel evidence about imprecise probabilities in a nationally-representative sample. We show, in a specific framework, that failing to account for imprecise or rounded probabilities can yield incorrect predictions of long-term care insurance purchase decisions.

We elicit numerical expectations for late-onset dementia and long-term-care (LTC) outcomes in the US Health and Retirement Study. We provide the first empirical evidence on dementia-risk perceptions among dementia-free older Americans and establish important patterns regarding imprecision of subjective probabilities. Our elicitation distinguishes between precise and imprecise probabilities, while accounting for rounding of reports. Imprecise-probability respondents quantify imprecision using probability intervals. Nearly half of respondents hold imprecise dementia and LTC probabilities, while almost a third of precise-probability respondents round their reports. These proportions decrease substantially when LTC expectations are conditioned on hypothetical knowledge of the dementia state. Among rounding and imprecise-probability respondents, our elicitation yields two measures: an initial rounded or approximated response and a post-probe response, which we interpret as the respondent's true point or interval probability. We study the mapping between the two measures and find that respondents initially tend to over-report small probabilities and under-report large probabilities. Using a specific framework for study of LTC insurance choice with uncertain dementia state, we illustrate the dangers of ignoring imprecise or rounded probabilities for modeling and prediction of insurance demand.

This special issue contains a selection of articles from the Second International Symposium on Imprecise Probabilities and Their Applications (ISIPTA ’01) [4]. The symposium was held at Cornell University, USA, in June 2001, and was chaired by Gert de Cooman, Terrence L. Fine, and Teddy Seidenfeld. It was a successor of the first successful ISIPTA meeting [3]. This was held in 1999, and was intended to promote interaction and communication between researchers in a great diversity of fields but with common interests in imprecise probabilities. ISIPTA ’01 was no exception in this respect: the 47 papers presented there covered a wide and heterogeneous range of topics, such as algorithms, belief change, civil engineering, pattern recognition, coherence, combination of uncertainties, conditioning, credal nets, decision making, economics and finance, elicitation, foundations, independence, mathematical models for uncertainty, probabilistic logic, reliability, and statistical inference. This heterogeneity as well as lively discussions during the meeting, turned ISIPTA ’01 into successful attempt to bring together an interdisciplinary community of researchers actively involved in advancing and promoting a generalized view of probability theories. Shortly after ISIPTA ’01, in February 2002, a few active members in this community decided to found SIPTA, the Society for Imprecise Probability: Theories and Applications. This society manages the organization of the ISIPTA meetings, nowadays a well-established international forum for discussing imprecise probabilities. And, being aware that the field of imprecise probabilities is not always easily accessible, due to its heterogeneity and the related variety of different Flanguages_ used to talk about imprecise probabilities, the Society has also recently started organizing the SIPTA Schools in Imprecise Probabilities, the next of which is due to be held in July 2006 in Madrid, Spain. At this point it is useful to be more precise about the subject of the ISIPTA meetings, namely, imprecise probabilities. The name Fimprecise probability_ was intended from the beginnings in a very wide sense, as a generic term for the many mathematical or statistical models that measure chance or uncertainty without sharp numerical probabilities. These models can be qualitative, such as comparative probability models and partial preference orderings, or quantitative, such as interval probabilities, belief functions, upper and lower expectations or previsions, just to name a few. More important, though, is that imprecise probability models are needed in inference problems characterized by scarce, vague or conflicting information. These

On grounds of insufficient reason, a probability of 1/K is assigned to K mutually exclusive possible events when nothing is known about the likelihood of those events. Fox and Rottenstreich (2003) present evidence that subjective probability judgments are typically biased towards this ignorance prior, and therefore depend on the partition K. Results from two studies indicate that lower-upper (imprecise) probability judgments by naive judges also exhibit partition dependence, despite the potential that imprecise probabilities provide for avoiding it. However, beta regression reveals two kinds of priming effects, one of which is modeled by mixture distributions. Another novel finding suggests that when partition primes conflict with a normatively correct partition some judges widen their probability intervals to encompass both partitions. The results indicate that imprecise probability judgments may be better suited than precise probabilities for handling conflicting or ambiguous information about ...

Description of the subject. The implementation of sensor based decision support in commercial dairy herds is highly dependent on having reliable systems. Problems with sensors give missing and noisy data hampering their use. Also, the presentation of results needs to be in a form which is simple and useful. These issues were addressed using a mastitis sensor and decision support as example. Objectives. This study aims at providing and evaluating a modular system applicable to the pipeline from sensor to decision support. Method. The case of mastitis was chosen as it is of economic importance and also affects welfare of cows, and because we have worked with a commercial sensor. The problems with sensors causing missing data and noise are described and a range of filtering and monitoring modules are shown to be important to make systems functional for herd management purposes. On top of this a solid method needs to be used to interpret and present data to end users, in terms of easy to read categories. Results. Filtering and pre-adjustments of raw data are important in making algorithms robust and reliable for daily use. Re-definition of traits is needed going from traditional few groups to continuous definitions, and then to new action oriented health classes. Also, for this case focusing on mastitis, assignment to “permanently sick” groups can be helpful in keeping focus on new acute cases. Conclusions. The combined use of filtering, fix-up routines and time series models leading into action oriented categories is needed to provide simple and robust decision support. The systems may be vastly improved by opening for transmission of data between user groups and to common databases – also with a few to use data in genetic selection.

In the area of preventive conservation special climate requirements are present. Especially fluctuations in climate values should be reduced in order to avoid damages of the sensitive materials of cultural heritage. In several applications no modern ventilation systems are present, such that the only ventilation option is the opening of windows and doors by human. Therefore serious climate fluctuations occur, if the ventilation strategy is not adapted to the climate situation. To avoid this situation, a monitoring and decision support system has to be developed. In this contribution we describe a fuzzy concept for such a system, analyze the influence of the fuzzy approach and explain the practical realization in a demonstrator application.

PurposeThis paper aims to introduce a practical expert decision support system (EDSS) for performing location analysis and making real estate location decisions in the organization’s facility and real estate management (FREM) department in presence of several decision criteria, under risk and uncertainty. This tool is particularly useful for making strategic decisions in facility planning, portfolio management, investment appraisal, development project evaluations and deciding on usage possibilities in an unbiased, objective manner.Design/methodology/approachThe proposed EDSS uses fuzzy logic and uncertainty theory as two of the most useful tools to deal with uncertainties involved in the problem environment. The system performs an unbiased mathematical analysis on the input data provided by the decision-maker, using a combination of Analytical Hierarchy Process (AHP) and Global Criterion Method; determines a suitable compromise level between the objectives; and delivers a set of locations that complies best with the outlined desires of the management as the final solution. The application of the system is tested on a real case and has delivered satisfactory results.FindingsThe proposed EDSS took the defined objectives, the list of alternative locations, and their attributes as the required input for problem-solving, and used a combination of AHP, Possibilistic approach, and global criterion method to solve the problem. The delivered outcome was a set of proper locations with the right attributes to meet all objectives of the organization at a satisfactory level, confirmed by the problem owners.Originality/valueThe application of such a system with such a degree of preciseness and complexity has been very limited in the literature. The system designed in this study is an Industry 4.0 decision making tool. For designing this system several body of knowledge are used. The present study is particularly useful for making strategic decisions in the domains of portfolio management, investment appraisal, project development evaluations and deciding on property usage possibilities. The proposed EDSS takes the information provided by the experts in the field (through qualitative and quantitative data collecting) as the inputs and operates as an objective decision-making tool using several bodies of knowledge considering the trends and developments in the world of FREM. The strong scientific method used in the core of the proposed EDSS guarantees a highly accurate result.

Poor planning and insufficient maintenance of existing structures and systems in developing countries like Nepal and Bangladesh have left them inadequate, unreliable, and vulnerable. Natural hazards have further exacerbated the impacts on these fragile, complex, and interconnected systems, which are expected to rise with the looming threat of increasing climate-induced uncertainties. Infrastructure resilience is, hence, indispensable for the sustainable development of any society. Therefore, a robust monitoring and decision support system is crucial in strengthening infrastructure resilience, which will inadvertently help in societal adaptation and resilience to hazards and threats. This paper discusses the effectiveness of digital technologies in reducing the risk of climate-induced hazards, taking the case of a climate change innovation – CRISTA developed by the Asian Institute of Technology and showcasing how the pilot solution was successful in addressing one of the pressing gaps in critical infrastructure risk management in Nepal. The CRISTA application collects real-time community information while delivering risk information and alerts. A human-AI hybrid network served as an interface between the data collected from real-time monitoring and the CRISTA dashboard. The vulnerability assessment for input in CRISTA was carried out using the IPCC AR5 risk assessment framework. This study also proves how digital technology applications can be beneficial for near-real-time monitoring, risk assessment, and decision-making in risk management of critical infrastructure systems by the use and integration of a crowdsourcing mobile application and associated geospatial dashboard.

The main goal of information fusion can be seen as improving human or automatic decision-making by exploiting diversities in information from multiple sources. High-level information fusion aims specifically at decision support regarding situations, often expressed as “achieving situation awareness”. A crucial issue for decision making based on such support is trust that can be defined as “accepted dependence”, where dependence or dependability is an overall term for many other concepts, e.g., reliability. This position paper reports on ongoing and planned research concerning imprecise probability as an approach to improved dependability in high-level information fusion. We elaborate on high-level information fusion from a generic perspective and a partial mapping from a taxonomy of dependability to high-level information fusion is presented. Three application domains: defense, manufacturing, and precision agriculture, where experiments are planned to be implemented are depicted. We conclude that high-level information fusion as an application-oriented research area, where precise probability (Bayesian theory) is commonly adopted, provides an excellent evaluation ground for imprecise probability.

Consequential life cycle assessment (CLCA) studies how a system responds to a decision in question. There has been a growing body of CLCA studies in the last decade, with different models being incorporated from other fields, partly to compensate for the limitations of the conventional linear models used in LCA. As much as we welcome the use of new models in (C)LCA, here we provide a cautionary note on this trend by highlighting the restrictiveness of assumptions underpinning different models. And we point to a path forward for future CLCA studies. We review the model setup of, and major assumptions behind, two classes of models used in CLCA studies. One is linear models such as process- or input-output-based LCA, which have been conventionally used in LCA. And the other is nonlinear optimization models such as computable general equilibrium (CGE), which are increasingly being applied in CLCA studies. While the linear models rest on several assumptions such as fixed coefficients and unlimited supply of inputs, so do the nonlinear optimization models. Among others, CGE models assume rationality, the limitations of which have been increasingly revealed by findings of experimental and behavioral economics. We also discuss some of the foundational questions. Are LCA estimates verifiable or falsifiable? If not, then is LCA science? And is the traditional definition of science based on falsifiability suited for LCA and other disciplines studying complex systems? Considering that (1) LCA studies the complex human environment system and model estimates or predictions are largely unverifiable and (2) different classes of models have different strengths and limitations, we make the following recommendations. For decision makers, particularly policy makers, we recommend evaluating estimates from different classes of models, as opposed to relying on a single class, for more robust decision support. Each model estimate or prediction can be taken as a point of evidence. If most estimates point to the same direction, the results would be considered strong evidence of what would happen. If, on the other hand, model estimates are scattered with no obvious patterns, the results would be considered inconclusive and thus more research is needed. For modelers, we recommend efforts be put into improving a model’s predictive capability by, e.g., relaxing some of the unrealistic assumptions such as fixed input/output coefficients, 1:1 perfect displacement, and systemic optimization. Our main message is that mathematical sophistication does not necessarily equal improvement in model accuracy. Given the complexity of the human - environment system, the uncertainties of predicting the future, and the limitations of different models, a multi-model approach is entailed for more robust decision-making, and continuous effort is needed to improve model predictability.

Approaches for extending logic to deal with uncertainty immanent to many real-world problems are often on the one side purely qualitative, such as modal logics, or on the other side quantitative, such as probabilistic logics. Research on combinations of qualitative and quantitative extensions to logic which put qualitative constraints on probability distributions, has mainly remained theoretical until now. In this paper, we propose a practically useful logic, which supports qualitative as well as quantitative uncertainty and can be extended with modalities with varying level of quantitative precision. This language has a solid semantic foundation based on imprecise probability theory. While in general imprecise probabilistic inference is much harder than the precise case, this is the first expressive imprecise probabilistic formalism for which probabilistic inference is shown to be as hard as corresponding precise probabilistic problems. A second contribution of this paper is an inference algorithm for this language based on the translation to a weighted model counting (WMC) problem, an approach also taken by state-of-the-art probabilistic inference methods for precise problems.

The SIPTA (Society for Imprecise Probability,Theory and Application) view of Keynes’s contributions to imprecise probability and application form a one-to-one, onto mapping from the claims of Henry E. Kyburg to what is accepted as being what Keynes’s contribution was. Thus, if one is familiar with Kyburg’s assessment of Keynes’s contributions, then one also knows what SIPTA’s views of Keynes’s contributions are. There is an extremely severe problem here, however, because Kyburg never read beyond chapter 6 in Keynes’s A Treatise on Probability. Kyburg was, therefore, ignorant about the many technical and mathematical contributions made to the theory of imprecise probability by Keynes, such as upper and lower probabilities, non additive probabilities,sub additive probabilities, interval valued estimates, decision weights, etc, in Parts II, III, IV and V of the A Treatise on Probability. Non additive probability appears in Keynes’s applications of non additive probability in his analysis of uncertainty in his theory of the liquidity preference approach to the rate of interest in the General Theory, where uncertainty was defined as an inverse function of the evidential weight of the argument in footnote 1 on page 148 of chapter 12. Keynes greatly emphasized the relative importance of his 'non-numerical' (interval) probabilities and weight of argument in his correspondence with Townshend in 1937-38. Kyburg’s view of Keynes’s contribution was to argue, based on Keynes’s verbal discussions on pp.30-34 of chapter III of Keynes’s A Treatise on Probability, that Keynes had been the first to (a) define a partial order and (b) the first to argue that all probabilities were not comparable and measurable. Kyburg also acknowledged Keynes’s contribution in chapter 6 concerning the weight of the argument as it related to the Bayesian refusal to give any importance to questions dealing with the relative strength of the evidence of different probabilities, i.e. whether the evidentiary base supporting a probability estimate was composed of strong evidence or made up of weak evidence. We will conclude that the above assessment made by Kyburg is identical to the SIPTA assessment made from 1999 (Kyburg’s 1999 assessment for SIPTA) to the last SIPTA meeting in 2019.

Updating beliefs with incomplete observations

Sensitivity analysis measures how changes in system inputs affect outputs. Previously, a large amount of sensitivity analysis research was relevant to the precise probability that is regarded as an ideal condition of engineering. Due to insufficient test samples and the low accuracy of test data, system reliability with hybrid uncertainty is difficult to be described as a precise value. As a profusion of highly integrated electromechanical equipment is applied in modern life, it is impossible to apply sufficient resources to eliminate the stochastic property of every component, which necessitates the identification of highly sensitive components to efficiently reduce imprecision. Hence, based on the theory of imprecise probability, imprecise sensitivity analysis has become a popular research topic in the last decade. In this paper, a method for uncertain system reliability and imprecise sensitivity analysis is proposed based on a Bayesian network, a probability box and the pinching method. The feasibility and accuracy of the combined method are fully verified through the evaluation and analysis of a numerical example and a case study of an electromechanical system, and the highly sensitive components that heavily influence the imprecision of system outputs are accurately identified.

This chapter offers a discussion of imprecision and indeterminacy in probability values; that is, there is a focus on situations where one does not attach a single real number to every possible event. There are several theories and mathematical models regarding such imprecise and indeterminate probabilities. Among these are, for instance, lower/upper or interval probabilities or expectations, Choquet capacities, belief functions and credal sets. The chapter reviews the history of some of these models and theories, and then summarizes their main technical assumptions. Some subtleties concerning conditioning and independence are also reviewed. The chapter concludes with a brief discussion of modeling and of criteria for decision-making (minimality, maximality, E-admissibility).