Finding the best national strategy to prevent or delay a country from acquiring nuclear weapons continues to be a critical issue for U.S. policy makers. In this paper, we build on previous work to develop a model that addresses this question. This model identifies the strategy that minimizes the disutility of the overall cost of the strategy and the cost of the consequences resulting from the strategy. We illustrate the insights that the model provides with a case study of Iran's nuclear weapons program.

We often make small ethical compromises for "good" reasons: We lie to a customer because our boss asked us to. We exaggerate our accomplishments on our resume to get an interview. Temptation blindsides us. And we make snap decisions we regret. Minor ethical lapses can seem harmless, but they instill in us a hard-to-break habit of distorted thinking. Rationalizations drown out our inner voice, and we make up the rules as we go. We lose control of our decisions, fall victim to the temptations and pressures of our situations, taint our characters, and sour business and personal relationships. In Ethics for the Real World, Ronald Howard and Clinton Korver explain how to master the art of ethical decision making by: Identifying potential compromises in your own life Applying distinctions to clarify your ethical thinking Committing in advance to ethical principles Generating creative alternatives to resolve dilemmas Packed with real-life examples, this book gives you practical advice to respond skillfully to life's inevitable ethical challenges. Not only can you make right decisions, you can acquire new habits that will realize the best in yourself and transform your relationships.

For centuries people have speculated on how to improve decision making without much professional help in developing clarity of action. Over the last several decades, many important contributions have been integrated to create the discipline of decision analysis, which can aid decision makers in all fields of endeavor: business, engineering, medicine, law, and personal life. Because uncertainty is the most important feature to consider in making decisions, the ability to represent knowledge in terms of probability, to see how to combine this knowledge with preferences in a reasoned way, to treat very large and complex decision problems using modern computation, and to avoid common errors of thought have combined to produce insights heretofore unobtainable. The limitation to its practice lies in our willingness to use reason, rather than in any shortcoming of the field. This chapter discusses the sources of the discipline; the qualities desired in a decision process; the logic for finding the best course of action; the process of focusing attention on important issues in attaining clarity of action; the need for clear and powerful distinctions to guide our thinking (because most decisions derive from thought and conversation rather than from computation); and the challenges to the growth of the discipline.

AboutSectionsView PDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked InEmail Go to Section HomeDecision AnalysisVol. 2, No. 3 Influence DiagramsRonald A. Howard, James E. MathesonRonald A. Howard, James E. MathesonPublished Online:1 Sep 2005https://doi.org/10.1287/deca.1050.0020 This article appears in INFORMS Analytics Collections Vol. 15: 25 Years of INFORMS. Visit this collection for free access to more articles showcasing the evolution of INFORMS over the past 25 years. Previous Back to Top Next FiguresReferencesRelatedInformationCited byIntermittent sampling for statistical process control with the number of defectivesComputers & Operations Research, Vol. 161A new way of measuring effects of financial crisis on contagion in currency marketsInternational Review of Financial Analysis, Vol. 90Who should I trust? Cautiously learning with unreliable experts27 September 2022 | Neural Computing and Applications, Vol. 35, No. 23Reasoning about causality in gamesArtificial Intelligence, Vol. 320Selecting cost-effective risk control option for advanced maritime operations; Integration of STPA-BN-Influence diagramOcean Engineering, Vol. 280Qualitative value of information provides a transparent and repeatable method for identifying critical uncertainty14 March 2023 | Ecological Applications, Vol. 33, No. 4Structured Decision Making to Prioritize Regional Bird Monitoring NeedsAuriel M. V. Fournier, R. Randy Wilson, Jeffrey S. Gleason, Evan M. Adams, Janell M. Brush, Robert J. Cooper, Stephen J. DeMaso, Melanie J. L. Driscoll, Peter C. Frederick, Patrick G. R. Jodice, Mary Ann Ottinger, David B. Reeves, Michael A. Seymour, Stephanie M. Sharuga, John M. Tirpak, William G. Vermillion, Theodore J. Zenzal, James E. Lyons, Mark S. Woodrey18 January 2023 | INFORMS Journal on Applied Analytics, Vol. 53, No. 3A Comprehensive Review of Data‐Driven Co‐Speech Gesture Generation23 May 2023 | Computer Graphics Forum, Vol. 42, No. 2Using Decision Analysis to Determine the Feasibility of a Conservation TranslocationLaura M. Keating, Lea Randall, Rebecca Stanton, Casey McCormack, Michael Lucid, Travis Seaborn, Sarah J. Converse, Stefano Canessa, Axel Moehrenschlager31 March 2023 | Decision Analysis, Vol. 0, No. 0Disaster world18 May 2022 | Computational and Mathematical Organization Theory, Vol. 29, No. 1Risk Assessment of Dike Based on Risk Chain Model and Fuzzy Influence Diagram28 December 2022 | Water, Vol. 15, No. 1Multi-Criteria Decision-Making Methods based on Fuzzy Sets10 January 2023The Co-design Process for Interactive Tools for Predicting Polygenic Risk Scores9 July 2023Thinking in systems: sustainable design of nano-enabled agriculture informed by life cycle assessmentImpact of Reliability in Conceptual Design—An Illustrative Trade-Off Analysis21 November 2022 | Systems, Vol. 10, No. 6Advancing construction of conditional probability tables of Bayesian networks with ranked nodes method5 July 2022 | International Journal of General Systems, Vol. 51, No. 8How to turn an MAS into a graphical causal model9 May 2022 | Autonomous Agents and Multi-Agent Systems, Vol. 36, No. 2Teaching Probabilistic Graphical Models with OpenMarkov30 September 2022 | Mathematics, Vol. 10, No. 19Supporting strategy selection in multiobjective decision problems under uncertainty and hidden requirementsEuropean Journal of Operational Research, Vol. 4Bayesian Decision Network–Based Optimal Selection of Hardening Strategies for Power Distribution SystemsASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, Vol. 8, No. 3Integrating Reliability in Conceptual Design Trade‐Off Analysis: A look at the Literature13 September 2022 | INCOSE International Symposium, Vol. 32, No. S2Application of qualitative value of information to prioritize uncertainties about eastern black rail population recovery25 May 2022 | Conservation Science and Practice, Vol. 4, No. 7Decision programming for mixed-integer multi-stage optimization under uncertaintyEuropean Journal of Operational Research, Vol. 299, No. 2A probabilistic framework for risk management and emergency decision-making of marine oil spill accidentsProcess Safety and Environmental Protection, Vol. 162Portraying probabilistic relationships of continuous nodes in Bayesian networks with ranked nodes methodDecision Support Systems, Vol. 154Scalable probabilistic estimates of electric vehicle charging given observed driver behaviorApplied Energy, Vol. 309Assessing Impacts of Health Threats and Health Security Actions7 June 2022Evaluating Health Threats and Health Security Actions Under Uncertainty7 June 2022Knowledge-Based Systems in Medicine10 November 2022On the Design of a Decision Support System for Robotic Equipment Adoption in Construction Processes2 December 2021 | Applied Sciences, Vol. 11, No. 23Prioritizing risk mitigation measures for binary heavy metal contamination emergencies at the watershed scale using bayesian decision networksJournal of Environmental Management, Vol. 299Resilience of Interdependent Water and Power Systems: A Literature Review and Conceptual Modeling Framework13 October 2021 | Water, Vol. 13, No. 20Statistical Process Control for the Number of Defectives with Limited MemoryBarry R. Cobb15 June 2021 | Decision Analysis, Vol. 18, No. 3Toward data-driven solutions to interactive dynamic influence diagrams8 August 2021 | Knowledge and Information Systems, Vol. 63, No. 9Risk of cardiovascular disease is driven by different combinations of environmental, medical and behavioral factors: Building a conceptual model for cumulative risk assessment26 May 2021 | Human and Ecological Risk Assessment: An International Journal, Vol. 27, No. 7Warning Decision-Making for Landslide Dam Breaching Flood Using Influence Diagrams6 August 2021 | Frontiers in Earth Science, Vol. 9Uncertainty representation in software models: a survey8 January 2021 | Software and Systems Modeling, Vol. 20, No. 4Norms in Counterfactual Selection8 June 2020 | Philosophy and Phenomenological Research, Vol. 103, No. 1Student Modelling on Language Teaching Based on Bayesian Networks27 June 2021 | Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım ve Teknoloji, Vol. 9, No. 2Predicting War Outcomes Based on a Fuzzy Influence Diagram6 February 2021 | International Journal of Fuzzy Systems, Vol. 23, No. 4Climate Risk ManagementAnnual Review of Earth and Planetary Sciences, Vol. 49, No. 1A computational framework for understanding the roles of simplicity and rational support in people's behavior explanationsCognition, Vol. 210Mapping conditional scenarios for knowledge structuring in (tail) dependence elicitation6 February 2020 | Journal of the Operational Research Society, Vol. 72, No. 4Four Decades of Transformation in Decision Analytic Practice for Societal Risk Management22 July 2019 | Risk Analysis, Vol. 41, No. 3Computing the value of information from periodic testing in holistic decision making under uncertaintyReliability Engineering & System Safety, Vol. 206Methods of Causal Analysis for Health Risk Assessment with Observational Data7 November 2020Using grounded theory and mental modeling to understand influences on electricians' safety decisions: Toward an integrated theory of why electricians work energizedSafety Science, Vol. 130Structured decision making in investigations involving digital and multimedia evidenceForensic Science International: Digital Investigation, Vol. 34Integer Programming on the Junction Tree Polytope for Influence DiagramsAxel Parmentier, Victor Cohen, Vincent Leclère, Guillaume Obozinski, Joseph Salmon24 July 2020 | INFORMS Journal on Optimization, Vol. 2, No. 3Analysis of the port sustainability parameters through Bayesian networksEnvironmental and Sustainability Indicators, Vol. 6A Discourse on the Incorporation of Organizational Factors into Probabilistic Risk Assessment: Key Questions and Categorical Review9 March 2020 | Risk Analysis, Vol. 40, No. 6A decision analysis approach to electronics standard development informed by life cycle assessment using influence diagramsJournal of Cleaner Production, Vol. 254A review of Gaussian Markov models for conditional independenceJournal of Statistical Planning and Inference, Vol. 206The influence of upflow velocity and hydraulic retention time changes on taxonomic and functional characterization in Fluidized Bed Reactor treating commercial laundry wastewater in co-digestion with domestic sewage7 April 2020 | Biodegradation, Vol. 31, No. 1-2ψ -epistemic interpretations of quantum theory have a measurement problem16 March 2020 | Quantum, Vol. 4Valuing portfolios of interdependent real options using influence diagrams and simulation-and-regression: A multi-stage stochastic integer programming approachComputers & Operations Research, Vol. 115International Journal of Disaster Risk Reduction, Vol. 47Decision support framework for bridge condition assessmentsStructural Safety, Vol. 81Value of Information in Probabilistic Logic Programs19 September 2019 | Electronic Proceedings in Theoretical Computer Science, Vol. 306Analysis of the Impact of Battlefield Environment on Military Operation Effectiveness Using Fuzzy Influence Diagram20 May 2019 | International Journal of Fuzzy Systems, Vol. 21, No. 6An introduction to Bayesian networks as assessment and decision support tools for managing coral reef ecosystem servicesOcean & Coastal Management, Vol. 177Using Set‐Based Design to Inform System Requirements and Evaluate Design Decisions27 September 2019 | INCOSE International Symposium, Vol. 29, No. 1Solving sequential collective decision problems under qualitative uncertaintyInternational Journal of Approximate Reasoning, Vol. 109GIRA: a general model for incident risk analysis11 September 2017 | Journal of Risk Research, Vol. 22, No. 2Assessment of Small-Scale Wind Turbines to Meet High-Energy Demand in Mexico with Bayesian Decision Networks27 October 2019Introducing Quantum-Like Influence Diagrams for Violations of the Sure Thing Principle21 November 2019Identifying management-relevant research priorities for responding to disease-associated amphibian declinesGlobal Ecology and Conservation, Vol. 16Distribution Network Risk Assessment Using Multicriteria Fuzzy Influence DiagramAdvances in Fuzzy Systems, Vol. 2018Modernizing the Bradford Hill criteria for assessing causal relationships in observational data15 November 2018 | Critical Reviews in Toxicology, Vol. 48, No. 8Mental Models of Navigation Safety to Inform Risk Management Decisions: Case Study on the Houston Ship ChannelASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, Vol. 4, No. 3Autonomous agents modelling other agents: A comprehensive survey and open problemsArtificial Intelligence, Vol. 258Representing causal knowledge in environmental policy interventions: Advantages and opportunities for qualitative influence diagram applications22 February 2018 | Integrated Environmental Assessment and Management, Vol. 14, No. 3An improved method for solving Hybrid Influence DiagramsInternational Journal of Approximate Reasoning, Vol. 95Learning Structures of Conceptual Models from Observed Dynamics Using Evolutionary Echo State Networks1 November 2017 | Journal of Artificial Intelligence and Soft Computing Research, Vol. 8, No. 2A spatial fuzzy influence diagram for modelling spatial objects' dependencies: a case study on tree-related electric outages11 October 2017 | International Journal of Geographical Information Science, Vol. 32, No. 2Robustness Metrics: How Are They Calculated, When Should They Be Used and Why Do They Give Different Results?6 February 2018 | Earth's Future, Vol. 6, No. 2The Selection of Experts for (Probabilistic) Expert Knowledge Elicitation18 November 2017Eliciting Multivariate Uncertainty from Experts: Considerations and Approaches Along the Expert Judgement Process18 November 2017An Information System Supporting the Eliciting of Expert Knowledge for Successful IT Projects29 August 2018A symbolic algebra for the computation of expected utilities in multiplicative influence diagrams21 June 2017 | Annals of Mathematics and Artificial Intelligence, Vol. 81, No. 3-4Decision AnalysisFuzzy evidential influence diagram and its evaluation algorithmKnowledge-Based Systems, Vol. 131An Approach to Evaluate the Profitability of Component Commonality18 May 2017 | Journal of Mechanical Design, Vol. 139, No. 7Directed Expected Utility NetworksManuele Leonelli, Jim Q. Smith9 May 2017 | Decision Analysis, Vol. 14, No. 2Is a time symmetric interpretation of quantum theory possible without retrocausality?21 June 2017 | Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 473, No. 2202An approach for reconciling different perspectives and stakeholder views on risk rankingJournal of Cleaner Production, Vol. 149Decision Analytic and Bayesian Uncertainty Quantification for Decision Support17 June 2017Predicting Corporate Credit Ratings Using Content Analysis of Annual Reports – A Naïve Bayesian Network Approach20 January 2017Use of expert knowledge to anticipate the future: Issues, analysis and directionsInternational Journal of Forecasting, Vol. 33, No. 1Preparing for the unprecedented — Towards quantitative oil risk assessment in the Arctic marine areasMarine Pollution Bulletin, Vol. 114, No. 1Bayesian Networks Improve Causal Environmental Assessments for Evidence-Based Policy8 December 2016 | Environmental Science & Technology, Vol. 50, No. 24Patterns in System Architecture Decisions25 January 2017 | Systems Engineering, Vol. 19, No. 6Analysis of decision alternatives of the deep borehole filter restoration problemEnergy, Vol. 114Improving ecosystem services modelling: Insights from a Bayesian network tools reviewEnvironmental Modelling & Software, Vol. 85MODELING THE NEW FRANCHISE CREATION DECISION: THE RELEVANCE OF BEHAVIORAL REASONS1 October 2016 | RAM. Revista de Administração Mackenzie, Vol. 17, No. 5ToPoMine: A graph miner for analysis of atom-dynamics simulation data in material scienceIntelligent Data Analysis, Vol. 20, No. 5Uncertainty reasoning for smart homes: An ontological decision network based approachWeb Intelligence, Vol. 14, No. 3Probabilistic Warnings in National Security Crises: Pearl Harbor RevisitedDavid M. Blum, M. Elisabeth Paté-Cornell8 December 2015 | Decision Analysis, Vol. 13, No. 1Representation, Reasoning, and Learning for a Relational Influence Diagram Applied to a Real-Time Geological Domain13 May 2016How Modeling Language Shapes Decisions: Problem-Theoretical Arguments and Illustration of an Example Case7 June 2016Using quantitative influence diagrams to map natural resource managers' mental models of invasive species managementLand Use Policy, Vol. 50Decision Analytic and Bayesian Uncertainty Quantification for Decision Support11 May 2016Mastering Situation Awareness3 December 2015 | ACM SIGMOD Record, Vol. 44, No. 3Adversarial Risk Analysis for Enhancing Combat Simulation Models23 November 2016 | Journal of Military Studies, Vol. 6, No. 2Robust decision making for UAV air-to-ground attack under severe uncertainty8 November 2015 | Journal of Central South University, Vol. 22, No. 11Model-Based Management of Strategic Initiatives4 July 2014 | Journal on Data Semantics, Vol. 4, No. 3A decision network account of reasoning about other people's choicesCognition, Vol. 142A probabilistic approach for a cost-benefit analysis of oil spill management under uncertainty: A Bayesian network model for the Gulf of FinlandJournal of Environmental Management, Vol. 158Evaluation of a Context-Aware Voice Interface for Ambient Assisted Living22 May 2015 | ACM Transactions on Accessible Computing, Vol. 7, No. 2Reliability model for helicopter main gearbox lubrication system using influence diagramsReliability Engineering & System Safety, Vol. 139Cognitive and Motivational Biases in Decision and Risk Analysis15 April 2015 | Risk Analysis, Vol. 35, No. 7Security risk analysis of system changes exemplified within the oil and gas domain1 October 2014 | International Journal on Software Tools for Technology Transfer, Vol. 17, No. 3Accurate lead time demand modeling and optimal inventory policies in continuous review systemsInternational Journal of Production Economics, Vol. 163Helicopter MGB oil system failure analysis using influence diagrams and random failure probabilitiesEngineering Failure Analysis, Vol. 50Collaborative Decision MakingDaniel Owen12 January 2015 | Decision Analysis, Vol. 12, No. 1Stress Testing Strategic Goals with SWOT Analysis8 December 2015The Concepts of Decision Making: An Analysis of Classical Approaches and Avenues for the Field of Enterprise Modeling26 November 2015Quantitative and Systemic Methods for Modeling SustainabilityEvolution of Security Engineering ArtifactsMulti-criteria Influence Diagrams – A Tool for the Sequential Group Risk AssessmentDecision-network polynomials and the sensitivity of decision-support modelsEuropean Journal of Operational Research, Vol. 239, No. 2Bibliography31 October 2014Evolution of Security Engineering ArtifactsInternational Journal of Secure Software Engineering, Vol. 5, No. 4Players in a Decision29 September 2014A Bayesian network decision model for supporting the diagnosis of dementia, Alzheimer׳s disease and mild cognitive impairmentComputers in Biology and Medicine, Vol. 51Model-driven risk analysis of evolving critical infrastructures27 February 2013 | Journal of Ambient Intelligence and Humanized Computing, Vol. 5, No. 2Advertising Media Selection and Delivery Decision-Making Using Influence Diagram5 December 2013Integrated Enterprise Risk Management: From Process to Best PracticeModern Economy, Vol. 05, No. 04Enhanced Adaptive Management: Integrating Decision Analysis, Scenario Analysis and Environmental Modeling for the Everglades11 October 2013 | Scientific Reports, Vol. 3, No. 1Modelling Vulnerability of Transportation Network Using Influence DiagramsCommunications - Scientific letters of the University of Zilina, Vol. 15, No. 4Cognitive mapping: A method to elucidate and present farmers' risk perceptionAgricultural Systems, Vol. 122Simulation metamodelling with Bayesian networks19 December 2017 | Journal of Simulation, Vol. 7, No. 4Co-speech Gesture Generation for Embodied Agents and its Effects on User Evaluation29 October 2013Climate change risk management: a Mental Modeling application3 September 2013 | Environment Systems and Decisions, Vol. 33, No. 3Modelling the risk-benefit impact of H1N1 influenza vaccines11 February 2013 | The European Journal of Public Health, Vol. 23, No. 4Bayesian Decision‐Network Modeling of Multiple Stakeholders for Reef Ecosystem Restoration in the Coral Triangle26 March 2013 | Conservation Biology, Vol. 27, No. 3LinkIT : A Ludic Elicitation Game for Eliciting Risk Perceptions18 October 2012 | Risk Analysis, Vol. 33, No. 6Decision-theoretic reflections on processing a fingermarkForensic Science International, Vol. 226, No. 1-3Appendix B: Influence Diagrams12 February 2013Perform Deterministic Analysis and Develop Insights12 February 2013Influence Diagrams23 January 2016Qualitative Analysis for Trust ManagementModeling decision-making biasesBiologically Inspired Cognitive Architectures, Vol. 3Value of information and mobility constraints for sampling with mobile sensorsComputers & Geosciences, Vol. 49Recent advances in applying decision science to managing national forestsForest Ecology and Management, Vol. 285An Inventory Criticality Classification Method for Nuclear Spare Parts25 July 2012Modeling and simulation of stranded passengers' transferring decision‐making on the basis of herd behaviorsKybernetes, Vol. 41, No. 7/8Cognitive Mapping Tools: Review and Risk Management Needs16 February 2012 | Risk Analysis, Vol. 32, No. 8Flood Risk Management: US Army Corps of Engineers and Layperson Perceptions9 May 2012 | Risk Analysis, Vol. 32, No. 8An evolutionary Bayesian belief network methodology for participatory decision making under uncertainty: An application to groundwater management23 June 2011 | Integrated Environmental Assessment and Management, Vol. 8, No. 3From the Editors—Games and Decisions in Reliability and RiskJason R. W. Merrick, Fabrizio Ruggeri, Refik Soyer, L. Robin Keller, 1 June 2012 | Decision Analysis, Vol. 9, No. 2Cognitive Networks Autonomic Decision-making Approach Based on Influence Diagram1 May 2012 | Journal of Networks, Vol. 7, No. 5From the Editor—Decisions over Time (Exploding Offers or Purchase Regret), in Game Settings (Embedded Nash Bargaining or Adversarial Games), and in Influence DiagramsL. Robin Keller, 1 March 2012 | Decision Analysis, Vol. 9, No. 1A Framework for Solving Hybrid Influence Diagrams Containing Deterministic Conditional DistributionsYijing Li, Prakash P. Shenoy, 1 March 2012 | Decision Analysis, Vol. 9, No. 1Individualized Gesture Production in Embodied Conversational Agents19 January 2012Integrated approach for the optimal selection of environmentally friendly drilling systemsInternational Journal of Energy and Environmental Engineering, Vol. 3, No. 1On-Line Technological Roadmapping as a Tool to Implement Foresight Results in IT EnterprisesCausality for Risk Analysts: Improving Our Understanding of How the World Works15 November 2012Environmental engineering of navigation infrastructure: A survey of existing practices, challenges, and potential opportunities19 September 2011 | Integrated Environmental Assessment and Management, Vol. 8, No. 1On eliciting requirements from end-users in the ICT4D domain14 July 2011 | Requirements Engineering, Vol. 16, No. 4Simulation metamodeling with dynamic Bayesian networksEuropean Journal of Operational Research, Vol. 214, No. 3Development of agent system based on decision model for creating an ambient space9 December 2010 | AI & SOCIETY, Vol. 26, No. 3A review of representation issues and modeling challenges with influence diagramsOmega, Vol. 39, No. 3Developing organisational decision‐making capability: a knowledge manager's guideJournal of Knowledge Management, Vol. 15, No. 3Sequential decision making with partially ordered preferencesArtificial Intelligence, Vol. 175, No. 7-8A network of options: Evaluating complex interdependent decisions under uncertaintyJournal of Economic Dynamics and Control, Vol. 35, No. 5Investment and Defense Strategies, Heuristics, and Games: From the Editor …L. Robin Keller, 1 March 2011 | Decision Analysis, Vol. 8, No. 1Towards the Applied Hybrid Model in Decision Making: A Neuropsychological Diagnosis of Alzheimer's Disease Study CaseInternational Journal of Computational Intelligence Systems, Vol. 4, No. 1Designing a Decision Support System20 July 2011Relating CORAS to the State of the ArtTask Analysis for Behavioral Factors Evaluation in Work System DesignInventory Management with Dynamic Bayesian Network Software SystemsFinancial Assessment of London Plan Policy 4A.2 by Probabilistic Inference and Influence DiagramsRonald A. Howard24 February 2011Return-to-Play in Sport: A Decision-based ModelClinical Journal of Sport Medicine, Vol. 20, No. 5MODELING THE PRODUCTION OF COVERBAL ICONIC GESTURES BY LEARNING BAYESIAN DECISION NETWORKSApplied Artificial Intelligence, Vol. 24, No. 6Structuring the decision problem8 July 2010Systematicity and Idiosyncrasy in Iconic Gesture Use: Empirical Analysis and Computational ModelingAn Interview with Ronald A. HowardRuss Garber, 6 November 2009 | Decision Analysis, Vol. 6, No. 4From the Editors …L. Robin Keller, Manel Baucells, John C. Butler, Philippe Delquié, Jason R. W. Merrick, Gregory S. Parnell, Ahti Salo, 1 December 2009 | Decision Analysis, Vol. 6, No. 4From the Editor…L. Robin Keller, 1 September 2009 | Decision Analysis, Vol. 6, No. 3Risk influence modeling of recent developments in helicopter safety on the Norwegian continental shelfGraph transformation based reduction analysis of PID6 July 2009 | ACM SIGSOFT Software Engineering Notes, Vol. 34, No. 4Theoretical tools for understanding and aiding dynamic decision makingJournal of Mathematical Psychology, Vol. 53, No. 3A Study of Pre-Decision Evaluation Using Influence Diagram: An Estimation of the Benefits of Influenza Vaccination25 June 2008 | Journal of Medical Systems, Vol. 33, No. 1Evaluating Situation Awareness of Autonomous Systems28 August 2009GNetIc – Using Bayesian Decision Networks for Iconic Gesture GenerationTowards the Neuropsychological Diagnosis of Alzheimer's Disease: A Hybrid Model in Decision MakingMarkov Logic: An Interface Layer for Artificial IntelligenceSynthesis Lectures on Artificial Intelligence and Machine Learning, Vol. 3, No. 1Decision Analysis15 September 2008From the Editor…L. Robin Keller, 1 September 2007 | Decision Analysis, Vol. 4, No. 3The Foundations of Decision Analysis RevisitedEnterprise architecture analysis with extended influence diagrams12 May 2007 | Information Systems Frontiers, Vol. 9, No. 2-3Decision strategies and design of agent interactions in hierarchical manufacturing systemsJournal of Manufacturing Systems, Vol. 26, No. 2A Diagnostic Tree for Improving Production Line Performance5 January 2009 | Production and Operations Management, Vol. 16, No. 1Modeling Air Combat by a Moving Horizon Influence Diagram GameJournal of Guidance, Control, and Dynamics, Vol. 29, No. 5Functional block diagrams and automated construction of event treesReliability Engineering & System Safety, Vol. 61, No. 3Energy Security for the Baltic RegionEvolutionary Bayesian Belief Networks for Participatory Water Resources Management under Uncertainty Volume 2, Issue 3September 2005Pages 125-181 Article Information Metrics Information Published Online:September 01, 2005 © 2005 INFORMSCite asRonald A. Howard, James E. Matheson, (2005) Influence Diagrams. Decision Analysis 2(3):127-143. https://doi.org/10.1287/deca.1050.0020 Keywordsinfluence diagramexpansionexpansion orderdecision treedecision-tree orderBayesarrow reversaldecision networkdecision-tree networkvalue of clairvoyanceBayesian networkbelief networkknowledge mapPDF download

Since the invention of Influence diagrams in the mid-1970s, they have become a ubiquitous tool for representing uncertain situations. This single diagram replaced awkward manipulations of decision trees and nature's trees with a single representation that displays both the sequential and informational structure of decisions. The diagram permits high-level graphic communication, clear assessments and computation in a single graphical system. This retrospective discusses the evolution and application of influence diagrams.

Ed Jaynes's view of probability, brilliantly clear, iconoclastic to many and still not sufficiently appreciated can transform thinking and lives. My professional life shows his seminal influence. I discuss his admiration of Laplace, and both the difficulty and joy of finding original references. Jayne's defense of and original work on Laplace's law of succession demonstrates how he has followed in the footsteps of his idol. Finally, Jaynes's theoretical and experimental investigations into Bertrand's Paradox illustrate the fundamental nature of his thought. Our loss of his presence is only compensated by the appreciation of his contribution.

We present an analogy between joint cumulative probability distributions and a class of multiattribute utility functions, which we call attribute dominance utility functions. Attribute dominance utility functions permit assessing multiattribute utility functions using common techniques of joint probability assessment such as marginal-conditional assessments and the method of copulas. By itself, this class of utility functions appears in many cases of decision analysis practice. Furthermore, we show that many functional forms of multiattribute utility function can be decomposed into attribute dominance utility functions that are easier to elicit. We introduce the notion of utility inference analogous to Bayes rule for probability inference and provide a graphic representation of attribute dominance utility functions, which we call utility diagrams.

In order to improve forecasts, a decision-maker often combines probabilities given by various sources, such as human experts and machine-learning classifiers. When few training data are available, aggregation can be improved by incorporating prior knowledge about both the event being forecasted and salient properties of the experts. To this end, we develop a generative Bayesian aggregation model for probabilistic classification of both binomial and multinomial events. The model includes an event-specific prior distribution, measures of individual experts' bias, calibration, and accuracy, and a measure of dependence between experts. Rather than require absolute measures, we show that the aggregate forecast may be expressed in terms of relative accuracy between experts. The model results in a weighted logarithmic opinion pool (LogOps) that satisfies the external Bayesian property, as well as a related consistency property that we term “invariance to indistinguishable outcomes.” We derive analytic solutions for the special cases of independent experts and for exchangeable experts. The model's application to decision analysis is demonstrated in a betting problem, calculating the value of experts as a function of their accuracy. We develop methods for assessing the model's parameters, as well as for learning them from data. Testing the model on simulated and real-world data demonstrates the model's accuracy to be comparable to or better than other aggregation methods.

This paper presents the current state of evolution of a language for teaching and practicing decision analysis that may avoid confusion of students, clients, and ourselves. Many of the terms currently used are inaccurate, arcane, or unnecessary. Restricting decision language to terms that are accurate, familiar, and fundamental contributes to clarity of thought and understanding. To illustrate the type of changes advocated, I propose replacingdependence withrelevance,expectation withe-value,utility withu-value, and eschewingsubjective probability,confidence,uncertainty about probability, any distinction betweenrisk and uncertainty, andstates of nature. I show how to incorporate the new terminology in teaching and practice.

These remarks respond to the comments offered by four colleagues on my paper "Speaking of Decisions: Precise Decision Language" (Howard 2004). The response gives me the opportunity to clarify and expand on the paper and to note different views. In particular, I further discussrandom variable, distinction, expected value, option, utility as a measure of happiness or satisfaction,subjective probability, the necessity of avalue measure for computing thevalue of clairvoyance, decision analysis as anormative discipline, and the inability of the clairvoyant to provide probabilities. I conclude with aconcept map for the language and a table showing where the terms are most useful in conversations with students and clients.

R a European mathematician who had spent many years on the theoretical study of Markov decision processes visited me and inquired about the range of application. I replied that I knew of very few practical applications and that I had found only one in the course of my career that I considered really successful. He was somewhat taken aback until we discussed the problem of the very large data requirements imposed by the Markov decision process formulation. Of course, he then asked me how these data requirements were met in my one successful application, and we spent the balance of his visit in discussing this application. I was reminded of this episode when Dr. Martin Puterman asked me a few months ago if I would comment on the origins of Markov decision processes. It occurred to me that the story of my one successful application might suit his purposes and be as interesting to the conference participants as it was to my European visitor. You see, my one successful application was the original application that sparked my interest in this whole research area. The story begins about twenty years ago, when I was a graduate student in electrical engineering at M.I.T., working part time with the Operations Research Group of Arthur D. Little, Inc. It was a period of great excitement in the Operations Research Group, for the first commercial applications were being performed with success, and there was considerable intellectual stimulation arising from many discussions of the fundamentals of the subject. Our mentor was Dr. George Kimball, who taught me and many of my colleagues the most valuable lessons I ever learned on the subject of operations research. The group had an important relationship with Sears, Roebuck and Company and helped that firm with a number of its problems. When I became involved, Sears’ management was becoming increasingly concerned with the effectiveness of the operation by which they sent catalogs to present and prospective customers. There were many options. A customer could receive up to fourteen individual mailings a year, ranging from the general catalog to individual sales fliers. Or, he might receive any subset of these mailings. The cost of any particular mailing was easily determined, but what was the benefit? To answer that question, let me take you back with me to a grey Chicago day, twenty years ago, when I first saw the Sears’ catalog information system. It was an unforgettable sight. Imagine, if you will, two or three acres of green steel filing cabinets. Each cabinet contained steel Addressograph plates, about four inches square. Each plate had a stencil for printing the customer’s name and address, and, as inserts, three small cards with several punched holes. These holes provided a limited summary over three seasons of the customer’s purchasing history as a Sears mail order client. About 100 young women continually circulated among the filing cabinets. They were supplied with one copy of the customer’s latest mail order, and it was their job to update the punches on the cards to incorporate the effect of the order. The information recorded was highly quantized in terms of the number and amount of the orders. The key to the system was the machine that used the steel plates to print labels for the catalog to be mailed. A drawer of plates was stacked into the machine. The machine examined the punched holes in the cards on each plate and determined, according to wired-board logic, whether this pattern of holes qualified the customer to receive the particular catalog being distributed at that time. If the pattern was favorable, a label was printed; otherwise, not. Thus, for example, the management could decide to send the general catalog only to customers who purchased more than $20 during the present season. By wiring the machine appropriately, this decision was implemented simply by passing every drawer of cards through the labeling machine. In making this decision, the management had traditionally looked at the direct profit to be expected from a customer as measured by the difference between the marginal profit on his purchases and the cost of sending him the catalog. As we examined the operation, we began to wonder whether it might be profitable to send catalogs, not just on the basis of the profit they might produce during the present season, but also for the impact they might have in moving the customer into more profitable categories in the future. We decided to model this system by what has come to be known as a Markovian decision process. Each customer’s state was described by his purchase history; there were a

A profession, since it affects the lives of others, must have ethics. Actions can be classified as prudential, legal, and ethical. Ethical theories are action based or consequence based; ethics can be positive, requiring action, or negative, proscribing action. Some people may have hierarchies establishing precedence among ethical rules. While harming and stealing are vital ethical concerns, most ethical issues in personal or professional life involve truth telling. Telling the whole truth requires going beyond not lying to a proactive concern with avoiding any deception. Keeping secrets is ethically sensitive because it changes your relationship with those who might benefit from knowing them. Issues of truth telling arise in business, academia, and in the lives of OR/MS professionals. Ethical dilemmas are best avoided by joining only those organizations whose ethical codes are consistent with your own, by refusing to participate in ethically objectionable activities, and by treating everyone as you would treat those you care about.

Business ethics, like other ethics, is usefully discussed by distinguishing prudential from legal from ethical actions. Prudential actions are those of simple self‐interest; legal actions are those not forbidden by the system for the use of force in society; ethical actions are those which you consider to be right. While ethics about physically not hurting people or not stealing from them are fundamental, most business ethics issues encountered revolve around truth telling. The ethic of not lying is insufficient: it still permits leaving false impressions, whether deliberate or not. A stronger, more satisfying ethic is “tell the truth”: fully inform the person with whom you are dealing. Truth telling is hard work for, often, we must learn the truth for ourselves before we can tell it to others. Truth telling may or may not lead to higher profits, but it will lead to a life without ethical remorse.

Over the past quarter century decision analysis has helped decision makers achieve clarity of action in fields that range from business to medicine. Some consulting companies practice primarily within the decision analysis paradigm. Major corporations strive to integrate decision analysis into their decision processes. Yet there are dissenters. To use a religious metaphor, heathens are those who pursue decision-aiding procedures that have different underlying philosophies. Heretics follow the general ideas of decision analysis but wish to change some underpinnings in a ruinous desire to achieve greater consistency with certain experiments involving unaided decision makers. Cults are composed of those who superficially follow the paradigm, but are willing to bend their practice in ways that allow the decision maker to avoid the dictates of logic while appearing to have done a decision analysis. Our examination of these views is based on the warranties each approach can support. The common sense warrantly requires, for example, that the addition of a noninformative new alternative cannot make a decision situation less desirable. A decision recommendation is no better than its warranties.

Let me begin by stating why I became an active participant in this discussion of the foundations of decision analysis. For many years, I had been blithely practicing in the field in the belief that all important questions about the underpinnings of the subject were now thoroughly understood and generally agreed upon. I knew that various theoreticians were developing alternate axiomatizations of decision theory, as they always had, but I did not believe that the resulting theories were being seriously proposed as bases for decision making. The onset of foreboding occurred when a prominent theorist made clear to me that he would reject a commonly accepted norm of decision analysis in making certain of his own decisions. I have previously described the conversation with the theorist that initiated my current militancy (Howard, 1988). However, I shall repeat it now for easy reference and because it so succinctly illustrates the issues at hand.

We examined allozyme variation in 10 breeding populations of Red-winged Blackbirds (Agelaius phoeniceus) across the United States, which represented samples from 9 of 14 putative subspecies in North America. Variation at 13 of 28 resolvable loci revealed a high level of genetic similarity for all seven populations from Florida and New York through the Great Plains to Oregon and northeastern California (pairwise Nei's distances, all less-than-or-equal-to 0.004). Differences in allozyme frequencies we found suggest that fewer subspecies exist in the continental United States than are currently recognized. The most interesting result was that the genetic distance between the populations sampled at Sacramento and San Francisco Bay national wildlife refuges, which are only 214 km apart, had a Nei's distance approximately 10 times as great as the genetic distance between Florida and Oregon populations. Salton Sea, California, the remaining population sampled, was also highly differentiated. Strong site fidelity, the nonmigratory behaviour of populations at Salton Sea and San Francisco, or both probably explain their relative allozyme distinctness, but the possibility that the brackish environment in which these birds live enforces a selective regime that reduces successful immigration or emigration to other habitats is intriguing.

Many would agree on the need to inform patients about the risks of medical conditions or treatments and to consider those risks in making medical decisions. The question is how to describe the risks and how to balance them with other factors in arriving at a decision. In this article, we present the thesis that part of the answer lies in defining an appropriate scale for risks that are often quite small. We propose that a convenient unit in which to measure most medical risks is the microprobability, a probability of 1 in 1 million. When the risk consequence is death, we can define a micromort as one microprobability of death. Medical risks can be placed in perspective by noting that we live in a society where people face about 270 micromorts per year from interactions with motor vehicles.Continuing risks or hazards, such as are posed by following unhealthful practices or by the side-effects of drugs, can be described in the same micromort framework. If the consequence is not death, but some other serious consequence like blindness or amputation, the microrisk structure can be used to characterize the probability of disability.Once the risks are described in the microrisk form, they can be evaluated in terms of the patient's willingness-to-pay to avoid them. The suggested procedure is illustrated in the case of a woman facing a cranial arteriogram of a suspected arterio-venous malformation. Generic curves allow such analyses to be performed approximately in terms of the patient's sex, age, and economic situation. More detailed analyses can be performed if desired.Microrisk analysis is based on the proposition that precision in language permits the soundness of thought that produces clarity of action and peace of mind.