DOCSLIB.ORG

Propagation of Uncertainties Modelled by Parametric P-Boxes Using Sparse Polynomial Chaos Expansions Roland Schöbi, Bruno Sudret

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Propagation of Uncertainties Modelled by Parametric P-Boxes Using Sparse Polynomial Chaos Expansions Roland Schöbi, Bruno Sudret Propagation of Uncertainties Modelled by Parametric P-boxes Using Sparse Polynomial Chaos Expansions Roland Schöbi, Bruno Sudret To cite this version: Roland Schöbi, Bruno Sudret. Propagation of Uncertainties Modelled by Parametric P-boxes Using Sparse Polynomial Chaos Expansions. 12th Int. Conf. on Applications of Statistics and Probability in Civil Engineering (ICASP12), Jul 2015, Vancouver, Canada. hal-01247151 HAL Id: hal-01247151 https://hal.archives-ouvertes.fr/hal-01247151 Submitted on 21 Dec 2015 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. 12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015 Propagation of Uncertainties Modelled by Parametric P-boxes Using Sparse Polynomial Chaos Expansions Roland Schöbi Ph.D. student, Chair of Risk, Safety and Uncertainty Quantification, ETH Zurich, Zurich, Switzerland Bruno Sudret Professor, Chair of Risk, Safety and Uncertainty Quantification, ETH Zurich, Zurich, Switzerland ABSTRACT: Advanced simulations, such as finite element methods, are routinely used to model the be- haviour of physical systems and processes. At the same time, awareness is growing on concepts of struc- tural reliability and robust design. This makes efficient quantification and propagation of uncertainties in computation models a key challenge. For this purpose, surrogate models, and especially Polynomial Chaos Expansions (PCE), have been used intensively in the last decade. In this paper we combine PCE and probability-boxes (p-boxes), which describe a mix of aleatory and epistemic uncertainty. In particu- lar, parametric p-boxes allow for separation of the latter uncertainties in the input space. The introduction of an augmented input space in PCE leads to a new uncertainty propagation algorithm for p-boxes. The proposed algorithm is illustrated with two applications: a benchmark analytical function and a realistic truss structure. The results show that the proposed algorithm is capable of predicting the p-box of the response quantity extremely efficiently compared to double-loop Monte Carlo simulation. 1. INTRODUCTION uncertainty), including probability-boxes (Ferson In modern engineering sciences, computational and Ginzburg, 1996), Bayesian hierarchical mod- simulations, such as finite element modelling, have els (Gelman, 2006) and Dempter-Shafer’s evidence become wide spread. The goal is to predict the re- theory (Dempster, 1967; Shafer, 1976). These sponse of a system with respect to a set of param- frameworks are generally referred to as imprecise eters, e.g. the deflection of a beam under variable probabilities. loads. The parameters (e.g. geometries, mechanical After the input uncertainty is characterized, properties, loads) are mapped to the quantity of in- it must be propagated through a computational terest through a computational model, e.g. through model. The latter, however, is often an expensive- the governing equations of the process. to-evaluate function, which can be replaced by an It is only in recent times that the traditionally approximate model, i.e. a meta-model, to reduce deterministic model parameters have been gradu- the computational effort needed. Well-known meta- ally substituted with probability distributions that modelling techniques include Polynomial Chaos account for their uncertainty. In practice though, Expansions (Ghanem and Spanos, 2003), Gaussian data available for calibrating such distributions are process modelling (a.k.a. Kriging) (Santner et al., often too sparse, thus resulting in an extra layer of 2003) and support vector machines (Gunn, 1998). uncertainty in their parameters. Different frame- This paper describes one formulation of impre- works have been proposed to quantify the latter cise probabilities in Section2 followed by an in- lack of knowledge (epistemic uncertainty) as well troduction to Polynomial Chaos expansions in Sec- as the natural variability of the process (aleatory tion3. Finally these two ingredients are combined 1 12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015 in Section4 and two applications are discussed in cannot be given a precise formulation. Thus the Section5. probability-box framework accounts for aleatory as well as for epistemic uncertainty in the description 2. INPUT UNCERTAINTY of a variable X. 2.1. Probability theory The lower and upper boundaries of the CDF are Traditionally, uncertainty in engineering has been denoted by [FX ;FX ]. The true, but unknown, CDF treated with probability theory. of X lies within the boundaries for any value of Consider a probability space (W;F ;P), where W x X, i.e. F (x) FX (x) FX (x); x X. The 2 X ≤ ≤ 8 2 denotes the event space equipped with s-algebra boundaries [FX ;FX ] mark the extreme cases of FX F and probability measure P. Random variables and are thus also CDFs by definition. are denoted by capital letters X(w) : W DX R The two boundaries form an intermediate space ! ⊂ where w W. A realization of variable X is de- in the variable-CDF-graph which resembles a box 2 noted by the corresponding lower case letters, e.g. (see Figure1), hence the name probability-box. x. Several random variables compose a random T 1 vector X = [X1;:::;XM] and the corresponding re- T alizations x = [x1;:::;xM] . P−box In this context a random variable X is described 0.8 cumulative distribution function F by its (CDF) X 0.6 Plausibility which expresses the probability that X < x, i.e. X F FX (x) = P(X < x). The CDF has the properties that 0.4 Belief it is monotone non-decreasing, that it tends to zero for low values of x (FX (x ¥) = 0) and that it 0.2 ! − tends to one for high values of x (FX (x ¥) = 1). ! 0 Note that such properties are valid for continuous −3 −2 −1 0 1 2 3 as well as discrete random variables. x For continuous random variables, the derivative Figure 1: P-box – CDF of a Gaussian variable with of a CDF is the probability density function fX (x) = interval-valued mX and sX dFX (x)=dx. The PDF describes the likelihood that X is in the neighbourhood of x. Due to the fact The p-box can be interpreted in the framework that the CDF is non-decreasing, the PDF has non- of Dempster-Shafer’s evidence theory (Dempster, negative values for all x X. 1967; Shafer, 1976). The lower boundary F de- 2 X As seen in the definitions above, probability the- scribes the minimum amount of probability that ory offers a single measure (i.e. the probability must be assigned to FX (x), which corresponds to the measure) to describe variability in variable X. In belief function Bel(FX (x)) in the vocabulary of ev- other words, we assume that the variability in X is idence theory. Analogously, the upper boundary of known and quantifiable by the CDF FX and the cor- the p-box is associated with the maximum amount responding PDF fX . This describes the case where of probability that might be assigned to FX (x), or variability is treated as the only source of uncer- the plausibility function Pl(FX (x)). tainty. Note that if FX (x) = FX (x) = FX (x); x X, then the p-box is called thin and conventional8 prob-2 2.2. Probability-box ability theory can be applied. A more general formulation is given by the frame- work of probability-boxes (p-boxes) which defines 2.3. Parametric p-boxes the CDF of a variable X by its lower and upper In the literature two types of p-boxes are identified, bound distributions (Ferson and Ginzburg, 1996; namely the free p-box and the parametric p-box. In Ferson and Hajagos, 2004). The idea is that due a this paper, we focus on parametric p-boxes (also lack of knowledge (epistemic uncertainty), the CDF called distributional p-boxes). 2 12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015 T A parametric p-box requires knowledge about The components of X = [X1;:::;XM] are assumed the shape of the true CDF but allows for uncertainty independent for the sake of simplicity throughout in its parameters. The p-box is represented by a this paper. The model response is a random vari- family of distribution functions whose parameters able Y obtained by propagating the input random q lie within an interval. For a single variable X: vector X through the computational model M . Several techniques are available for surrogating F (x) = F (x;q ); (1) X X the expensive-to-evaluate computational model M . In the following section, Polynomial Chaos Expan- where qi [q i;q i]; i = 1;:::;nq . This construction resembles2 a Bayesian hierarchical model (Gelman, sions (Ghanem and Spanos, 2003; Sudret, 2007) 2006) in which the distribution of the parameters will be briefly introduced. q is replaced by an interval. This framework al- 3.2. Polynomial Chaos Expansion lows for a clear separation of aleatory and epistemic A well-known non-intrusive meta-modelling uncertainty: aleatory uncertainty is represented by method is Polynomial Chaos Expansion (PCE) the distribution function family and epistemic un- which approximates the computational model certainty is represented by the interval on parame- M with a finite series of polynomials orthogonal with ters q . respect to the distribution of the input variables: Figure2 illustrates a parametric p-box generated by a Gaussian random variable with mean value (PCE) Y M (X ) = aa ya (X ); (2) and standard deviation varying within the intervals ≈ ∑ a A M;p 2 mX = [ 0:5;0:5] and sX = [0:7;1:0].
Load more

Recommended publications
  • Literature Review of Methods for Representing Uncertainty
    Methods for representing uncertainty A literature review Enrico Zio and Nicola Pedroni n° 2013-03 THEME Risk analysis he Foundation for an Industrial Safety Culture (FonCSI) is a french public-interest research T foundation created in 2005. It aims to: undertake and fund research activities that contribute to improving safety in hazardous organizations (industrial firms of all sizes, in all industrial sectors); work towards better mutual understanding between high-risk industries and civil society, aiming for a durable compromise and an open debate that covers all the dimensions of risk; foster the acculturation of all stakeholders to the questions, tradeoffs and problems related to risk and safety. In order to attain these objectives, the FonCSI works to bring together researchers from different scientific disciplines with other stakeholders concerned by industrial safety and the management of technological risk: companies, local government, trade unions, NGOs. We also attempt to build bridges between disciplines and to promote interaction and cross-pollination between engineering, sciences and the humanities. The work presented in this document is the result of research funded by the FonCSI.The opinions presented are those of the authors. Foundation for an Industrial Safety Culture A public-interest research foundation http://www.FonCSI.org/ 6 allée Émile Monso — BP 34038 Telephone: +33 534 32 32 00 31029 Toulouse cedex 4 Twitter: @LaFonCSI France Email: [email protected] iii Résumé Titre Revue de littérature sur la représentation de l’incertitude dans l’analyse de risque Mots-clefs incertitude, analyse de risque, approches possibilistes, analyse d’intervalle Auteurs Enrico Zio et Nicola Pedroni Date de publication avril 2013 Ce document fournit une analyse critique de différentes méthodes de représentation et d’analyse d’incertitude dans les analyses de risques: l’analyse probabiliste classique, les probabilités imprécises (analyse d’intervalle), les fonctions de croyance de Dempster-Shafer et la théorie des possibilités.
    [Show full text]
  • Uncertainty Modelling for Scarce and Imprecise Data in Engineering Applications
    Uncertainty Modelling for Scarce and Imprecise Data in Engineering Applications Thesis submitted in accordance with the requirements of the University of Liverpool for the degree of Doctor in Philosophy by Jonathan Cyrus Sadeghi January 2020 Uncertainty Modelling for Scarce and Imprecise Data in Engineering Applications Abstract In this thesis, models for uncertainty quantification in the case of scarce and imprecise data are described, and the computational efficiency of simulations with these models is improved. Specifically, probability boxes are used to describe imprecision in cumulative distribution functions. This may be the case when imprecise data is used to train a model, or the prior knowledge regarding a property of the system being studied is very weak. Performing simulations with probability boxes is often computationally expensive, because an optimisation program must be solved to obtain each sample in a Monte Carlo simulation. When the system model is known analytically, it is possible to significantly reduce the cost of the analysis. However, the system model is often a black box which can only be queried for a particular point value of the input. Each evaluation or query of the system model is often computationally expensive in itself. Currently, few efficient methods exist to perform computations with probability boxes, and the techniques which exist do not provide rigorous bounds on the obtained probability of failure. Interval Predictor Models are a technique to create an approximate representation of a function, where the uncertainty in the true function is described as an interval, with statistical guarantees on the coverage of the true function. This thesis proposes the use of Interval Predictor Models to create an approximate surrogate model for the true black box system model and hence obtain rigorous bounds on the probability of failure of a system.
    [Show full text]
  • Reliability of Computational Science
    Reliability of Computational Science I. Babuska∗ F. Nobile† R. Tempone‡ November 2, 2006 Abstract Today’s computers allow us to simulate large, complex physical prob- lems. Many times the mathematical models describing such problems are based on a relatively small amount of available information such as exper- imental measurements. The question arises whether the computed data could be used as the basis for decision in critical engineering, economic, medicine applications. The representative list of engineering accidents oc- curred in the past years and their reasons illustrates the question. The paper describes a general framework for Verification and Validation which deals with this question. The framework is then applied to an illustrative engineering problem, in which the basis for decision is a specific quantity of interest, namely the probability that the quantity does not exceed a given value. The V&V framework is applied and explained in detail. The result of the analysis is the computation of the failure probability as well as a quantification of the confidence in the computation, depending on the amount of available experimental data. 1 Introduction Computational Science is a discipline concerned with the use of computers for the prediction of physical phenomena. These predictions are used as the basis for critical decisions in engineering and in other fields such as environment, heath, management, etc. Rapid development of computer hardware allows us to make predictions of more and more complex phenomena. The major problem arises: How reliable are these predictions? Could they be the basis for decisions, often very crucial and with large implications? The reliability problem has many aspects: mathematical, numerical, computational, ∗ICES, University of Texas at Austin, Texas, USA.
    [Show full text]
  • Uncertainty Propagation of P-Boxes Using Sparse Polynomial Chaos Expansions R Schöbi, B Sudret
    Uncertainty propagation of p-boxes using sparse polynomial chaos expansions R Schöbi, B Sudret To cite this version: R Schöbi, B Sudret. Uncertainty propagation of p-boxes using sparse polynomial chaos expansions. 2016. hal-01421112 HAL Id: hal-01421112 https://hal.archives-ouvertes.fr/hal-01421112 Preprint submitted on 21 Dec 2016 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. UNCERTAINTY PROPAGATION OF P-BOXES USING SPARSE POLYNOMIAL CHAOS EXPANSIONS R. Schobi,¨ B. Sudret CHAIR OF RISK,SAFETY AND UNCERTAINTY QUANTIFICATION Risk, Safety & STEFANO-FRANSCINI-PLATZ 5 Uncertainty Quantification CH-8093 ZURICH¨ Data Sheet Journal: - Report Ref.: RSUQ-2016-009 Arxiv Ref.: http://arxiv.org/abs/1608.05565 - [stat.CO] DOI: - Date submitted: August 19, 2016 Date accepted: - Uncertainty propagation of p-boxes using sparse polynomial chaos expansions R. Sch¨obi1 and B. Sudret1 1Chair of Risk, Safety and Uncertainty Quantification, ETH Zurich, Stefano-Franscini-Platz 5, 8093 Zurich, Switzerland Abstract In modern engineering, physical processes are modelled and analysed using advanced com- puter simulations, such as finite element models. Furthermore, concepts of reliability analysis and robust design are becoming popular, hence, making efficient quantification and propagation of uncertainties an important aspect.
    [Show full text]
  • On the Calculation of the Bounds of Probability of Events Using Infinite
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector International Journal of Approximate Reasoning 43 (2006) 241–267 www.elsevier.com/locate/ijar On the calculation of the bounds of probability of events using infinite random sets Diego A. Alvarez * Arbeitsbereich fu¨r Technische Mathematik am Institut fu¨r Grundlagen der Bauingenieurwissenschaften, Leopold-Franzens Universita¨t, Technikerstrasse 13, A-6020 Innsbruck, EU, Austria Received 5 December 2005; received in revised form 9 March 2006; accepted 19 April 2006 Available online 22 May 2006 Abstract This paper presents an extension of the theory of finite random sets to infinite random sets, that is useful for estimating the bounds of probability of events, when there is both aleatory and epistemic uncertainty in the representation of the basic variables. In particular, the basic variables can be mod- elled as CDFs, probability boxes, possibility distributions or as families of intervals provided by experts. These four representations are special cases of an infinite random set. The method intro- duces a new geometrical representation of the space of basic variables, where many of the methods for the estimation of probabilities using Monte Carlo simulation can be employed. This method is an appropriate technique to model the bounds of the probability of failure of structural systems when there is parameter uncertainty in the representation of the basic variables. A benchmark example is used to demonstrate the advantages and differences of the proposed method compared with the finite approach. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Random sets; Dempster–Shafer evidence theory; Epistemic uncertainty; Aleatory uncertainty, Monte Carlo simulation * Tel.: +43 512 5076825; fax: +43 512 5072941.
    [Show full text]
  • A Critical Discussion and Practical Recommendations on Some Issues
    A critical discussion and practical recommendations on some issues relevant to the non-probabilistic treatment of uncertainty in engineering risk assessment Nicola Pedroni 1, Enrico Zio 2, Alberto Pasanisi 3, Mathieu Couplet 4 1Corresponding author. Chair “System Science and the Energy challenge”-Fondation Electricité de France (EdF) at the Laboratoire Genie Industriel (LGI), CentraleSupélec, Université Paris-Saclay, Grande voie des Vignes, 92290 Chatenay-Malabry, France. E-mail: [email protected] 2Chair “System Science and the Energy challenge”-Fondation Electricité de France (EdF) at the Laboratoire Genie Industriel (LGI), CentraleSupélec, Université Paris-Saclay, Grande voie des Vignes, 92290 Chatenay-Malabry, France. E-mail: [email protected] . Also: Energy Department, Politecnico di Milano, Via Ponzio, 34/3 – 20133 Milano, Italy. E-mail: [email protected] . 3EDF R&D - EIFER, Emmy-Noether-Str. 11. 76131 Karlsruhe, Germany. E-mail address: [email protected] . 4Electricité de France, R&D, 6 Quai Watier, 78400, Chatou, France. E-mail address: [email protected] . 1 ABSTRACT Models for the assessment of the risk of complex engineering systems are affected by uncertainties, due to the randomness of several phenomena involved and the incomplete knowledge about some of the characteristics of the system. The objective of the present paper is to provide operative guidelines to handle some conceptual and technical issues related to the treatment of uncertainty in risk assessment for engineering practice. In particular, the following issues are addressed: (1) quantitative modeling and representation of uncertainty coherently with the information available on the system of interest; (2) propagation of the uncertainty from the input(s) to the output(s) of the system model; (3) (Bayesian) updating as new information on the system becomes available; (4) modeling and representation of dependences among the input variables and parameters of the system model.
    [Show full text]
  • Belief Functions: a Gentle Introduction Seoul National University
    Belief functions: A gentle introduction Seoul National University Professor Fabio Cuzzolin School of Engineering, Computing and Mathematics Oxford Brookes University, Oxford, UK Seoul, Korea, 30/05/18 Professor Fabio Cuzzolin Belief functions: A gentle introduction Seoul, Korea, 30/05/18 1 / 125 Uncertainty Outline Belief functions Decision making 1 Uncertainty Semantics 5 Second-order uncertainty Dempster’s rule Theories of uncertainty Classical probability Multivariate analysis Imprecise probability 2 Beyond probability Misunderstandings Monotone capacities Probability intervals Set-valued observations 4 Reasoning with belief Propositional evidence Fuzzy and possibility theory functions Probability boxes Scarce data Statistical inference Rough sets Representing ignorance Combination Rare events Conditioning 6 Belief functions on reals Uncertain data Belief vs Bayesian reasoning Continuous belief functions 3 Belief theory Generalised Bayes Theorem Random sets A theory of evidence The total belief theorem 7 Conclusions Professor Fabio Cuzzolin Belief functions: A gentle introduction Seoul, Korea, 30/05/18 2 / 125 Uncertainty Second-order uncertainty Orders of uncertainty the difference between predictable and unpredictable variation is one of the fundamental issues in the philosophy of probability second order uncertainty: being uncertain about our very model of uncertainty has a consequence on human behaviour: people are averse to unpredictable variations (as in Ellsberg’s paradox) how good are Kolmogorov’s measure-theoretic probability,
    [Show full text]
  • CMIS Modelling Optimisation Simulation
    Uncertainty and uncertainty analysis methods Keith R. Hayes February 2011 Report Number: EP102467 Issues in quantitative and qualitative risk modeling with application to import risk assessment ACERA project (0705) Enquiries should be addressed to: Keith R Hayes CSIRO Mathematics, Informatics and Statistics GPO Box 1538 Hobart, Tasmania, Australia, 7001 Castray Esplanade, Hobart, Australia, 7000 Telephone : +61 3 6232 5260 Fax : +61 3 6232 5485 Email : [email protected] Distribution List Client (1) Publications Officer (X) Stream Leader (1) Authors (1) Copyright and Disclaimer c CSIRO To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO. Important Notice CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consul- tants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. Contents 1 Introduction..................................... 8 1.1 Project background...............................8 1.2 Report structure and outline...........................8 1.3 Fundamental issues..............................
    [Show full text]
  • Eastern Box Turtles (Terrapene C. Carolina) in the Oak Openings Region of Northwestern Ohio
    SPATIAL ECOLOGY OF EASTERN BOX TURTLES (TERRAPENE C. CAROLINA) IN THE OAK OPENINGS REGION OF NORTHWESTERN OHIO Matthew D. Cross A Dissertation Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY August 2016 Committee: Karen V. Root, Advisor Salim Elwazani Graduate Faculty Representative Enrique Gomezdelcampo Jeffery G. Miner Shannon L. Pelini © 2016 Matthew D. Cross All Rights Reserved iii ABSTRACT Karen V. Root, Advisor Eastern Box Turtles (Terrapene c. carolina) have experienced range-wide declines as the result of extensive habitat loss, fragmentation, and alteration. The Oak Openings Region of northwestern Ohio is a biodiversity hotspot that exists in a highly fragmented landscape and provides a unique case study from which to examine the effects of anthropogenic disturbance on Eastern Box Turtles. In an effort to inform management and conservation efforts in the Oak Openings Region, I initiated a radio-telemetry project with the following objectives: 1) examine the spatial ecology of Eastern Box Turtles over several years to understand how they interact with their habitat in an area as unique as the Oak Openings Region, 2) develop predictive models depicting the temporal distributions of Eastern Box Turtles, 3) examine the impacts of one of the most common management tools in the Oak Openings Region, prescribed fire, on Eastern Box Turtles, and 4) evaluate pattern-recognition software as a low-cost alternative of identifying individual Eastern Box Turtles. Turtles at my study site exhibited larger home ranges than previously reported for this species as well as hierarchical habitat selection at multiple scales.
    [Show full text]
  • Propagation of Imprecise Probabilities Through
    PROPAGATION OF IMPRECISE PROBABILITIES THROUGH BLACK-BOX MODELS A Thesis Presented to The Academic Faculty by Morgan Bruns In Partial Fulfillment of the Requirements for the Degree Master of Science in the School of Mechanical Engineering Georgia Institute of Technology May 2006 PROPAGATION OF IMPRECISE PROBABILITIES THROUGH BLACK-BOX MODELS Approved by: Dr. Christiaan J.J. Paredis, Advisor School of Mechanical Engineering Georgia Institute of Technology Dr. Bert Bras School of Mechanical Engineering Georgia Institute of Technology Dr. Leon F. McGinnis School of Industrial and Systems Engineering Georgia Institute of Technology Date Approved: April 7, 2006 ACKNOWLEDGEMENTS This thesis would not have been possible without the assistance of many of my friends and colleagues. I would like to thank my advisor, Dr. Chris Paredis, for his support, guidance, and insight. I would also like to thank the students of the Systems Realization Laboratory for the intellectually nurturing environment they have provided. In particular, I would like to thank Jason Aughenbaugh, Scott Duncan, Jay Ling, Rich Malak, and Steve Rekuc for their criticism and advice regarding my research. iii TABLE OF CONTENTS ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES vii LIST OF ABBREVIATIONS viii LIST OF SYMBOLS ix SUMMARY xi CHAPTER 1 : INTRODUCTION 1 1.1 Design decision making 1 1.2 Imprecision in design 2 1.3 Challenges for decision making formalisms 4 1.4 Research questions, hypotheses, and thesis outline 5 CHAPTER 2 : DECISION MAKING UNDER IMPRECISE UNCERTAINTY
    [Show full text]
  • SIPTA Summer School: Engineering
    SIPTA Summer School: Engineering Day Scott Ferson, Institute for Risk and Uncertainty, University of Liverpool Society for Imprecise Probabilities: Theory and Applications 24-28 July 2018, Oviedo, Spain © 2018 Scott Ferson Engineering • Publication advice joke • Practical applications • Get things done, even if not perfectly • Random variables, not just events • Logic, arithmetic, FE and ODE models • Continuous, infinite • Quick & dirty is better than elegant & abstract • Sloppy with notation, but big on computability Installing R • Locate the latest version by Googling "R" or going to http://www.r-project.org/ • Download and install the base software – For instance, get version 3.5.0 for Windows from https://cloud.r-project.org/bin/windows/base/R-3.5.0-win.exe • Install R by executing this self-extracting file and invoke the program by double-clicking on the blue R icon it leaves on your desktop Leaving R • To leave R click File / Exit from the main menu or enter the command quit() • R may ask whether you want to save your work • If you do, your variables and functions will be available the next time you invoke R Install the pba.r probability box library • If you left R, invoke it again • Enter rm(list = ls()) to clear its memory • Click File / Source R code on the main menu • Locate and Open the file pba.r • You’ll get the message ":pba> library loaded" Die-hard RStudio users • If, against advice, you must use RStudio… • File / Open file pbox.r • Source (Ctrl+Shift+S) pbox.r • Enter the instruction RStudio = TRUE pba.r probability bounds
    [Show full text]
  • Verified Computation with Probabilities
    Verified Computation with Probabilities Scott Ferson and Jack Siegrist Applied Biomathematics Abstract. Because machine calculations are prone to errors that can sometimes accumulate disastrously, computer scientists use special strategies called verified computation to ensure output is reliable. Such strategies are needed for computing with probability distributions. In probabilistic calculations, analysts have routinely assumed (i) probabil- ities and probability distributions are precisely specified, (ii) most or all variables are independent or otherwise have well-known dependence, and (iii) model structure is known perfectly. These assumptions are usually made for mathematical convenience, rather than with empirical justi- fication, even in sophisticated applications. Probability bounds analysis computes bounds guaranteed to enclose probabilities and probability dis- tributions even when these assumptions are relaxed or removed. In many cases, results are best-possible bounds, i.e., tightening them requires ad- ditional empirical information. This paper presents an overview of prob- ability bounds analysis as a computationally practical implementation of the theory of imprecise probabilities that represents verified computation of probabilities and distributions. Keywords: probability bounds analysis, probability box, p-box, veri- fied computation, imprecise probabilities, interval analysis, probabilistic arithmetic. 1 Introduction Many high-profile disasters are attributable to numerical errors in computer calculations. The self-destruction of the Ariane 5 rocket on its maiden test flight was caused by integer overflow (ESA 1996). The crash of the Mars Climate Orbiter during orbital insertion resulted from a units incompatibility (Isbell et al. 1999). The Sleipner A offshore drilling platform sank because of an inaccurate finite element approximation (Selby et al. 1997). The Aegis cruiser USS Yorktown was dead in the water for several hours because of a propagated divide-by-zero error (Slabodkin 1998).
    [Show full text]