Copyright © 2009 PMI RiskSIGNovember 5-6, 2009 RiskSIG - Advancing the State of the Art A collaboration of the PMI, Rome Italy Chapter and the RiskSIG.

Slides:



Advertisements
Similar presentations
A Tutorial on Learning with Bayesian Networks
Advertisements

1 Some Comments on Sebastiani et al Nature Genetics 37(4)2005.
Decision Making Under Risk Continued: Bayes’Theorem and Posterior Probabilities MGS Chapter 8 Slides 8c.
Psychology 290 Special Topics Study Course: Advanced Meta-analysis April 7, 2014.
Bayesian Estimation in MARK
Introduction of Probabilistic Reasoning and Bayesian Networks
Chapter 15 Probabilistic Reasoning over Time. Chapter 15, Sections 1-5 Outline Time and uncertainty Inference: ltering, prediction, smoothing Hidden Markov.
Uncertain knowledge and reasoning
1. Profile Decision-making and risk assessment under uncertainty Special expertise on software project risk assessment Novel applications of causal models.
Bayesian inference Gil McVean, Department of Statistics Monday 17 th November 2008.
Introduction  Bayesian methods are becoming very important in the cognitive sciences  Bayesian statistics is a framework for doing inference, in a principled.
1 Chapter 12 Probabilistic Reasoning and Bayesian Belief Networks.
Learning with Bayesian Networks David Heckerman Presented by Colin Rickert.
Part 3 of 3: Beliefs in Probabilistic Robotics. References and Sources of Figures Part 1: Stuart Russell and Peter Norvig, Artificial Intelligence, 2.
Bayesian Belief Networks
1 Learning Entity Specific Models Stefan Niculescu Carnegie Mellon University November, 2003.
AI - Week 24 Uncertain Reasoning (quick mention) then REVISION Lee McCluskey, room 2/07
Part 2 of 3: Bayesian Network and Dynamic Bayesian Network.
1 Bayesian Reasoning Chapter 13 CMSC 471 Adapted from slides by Tim Finin and Marie desJardins.
. PGM: Tirgul 10 Parameter Learning and Priors. 2 Why learning? Knowledge acquisition bottleneck u Knowledge acquisition is an expensive process u Often.
Uncertainty Logical approach problem: we do not always know complete truth about the environment Example: Leave(t) = leave for airport t minutes before.
Lecture 9: p-value functions and intro to Bayesian thinking Matthew Fox Advanced Epidemiology.
Quiz 4: Mean: 7.0/8.0 (= 88%) Median: 7.5/8.0 (= 94%)
Statistical Analysis of Systematic Errors and Small Signals Reinhard Schwienhorst University of Minnesota 10/26/99.
Project Management Chapter 13 OPS 370. Projects Project Management Five Phases 1. Initiation 2. Planning 3. Execution 4. Control 5. Closure.
The Bayesian Web Adding Reasoning with Uncertainty to the Semantic Web
1 Bayesian methods for parameter estimation and data assimilation with crop models Part 2: Likelihood function and prior distribution David Makowski and.
Graphical Causal Models: Determining Causes from Observations William Marsh Risk Assessment and Decision Analysis (RADAR) Computer Science.
Bayesian networks Chapter 14. Outline Syntax Semantics.
A Brief Introduction to Graphical Models
Soft Computing Lecture 17 Introduction to probabilistic reasoning. Bayesian nets. Markov models.
Digital Statisticians INST 4200 David J Stucki Spring 2015.
Renaissance Risk Changing the odds in your favour Risk forecasting & examples.
Bayesian Networks for Data Mining David Heckerman Microsoft Research (Data Mining and Knowledge Discovery 1, (1997))
Kansas State University Department of Computing and Information Sciences CIS 730: Introduction to Artificial Intelligence Lecture 28 of 41 Friday, 22 October.
CSE PR 1 Reasoning - Rule-based and Probabilistic Representing relations with predicate logic Limitations of predicate logic Representing relations.
Bayesian statistics Probabilities for everything.
why smart data is better than big data Queen Mary University of London
1 Chapter 12 Probabilistic Reasoning and Bayesian Belief Networks.
Screening of diseases Dr Zhian S Ramzi Screening 1 Dr. Zhian S Ramzi.
This material was developed by Oregon Health & Science University, funded by the Department of Health and Human Services, Office of the National Coordinator.
On Predictive Modeling for Claim Severity Paper in Spring 2005 CAS Forum Glenn Meyers ISO Innovative Analytics Predictive Modeling Seminar September 19,
4 Proposed Research Projects SmartHome – Encouraging patients with mild cognitive disabilities to use digital memory notebook for activities of daily living.
SOFTWARE METRICS Software Metrics :Roadmap Norman E Fenton and Martin Neil Presented by Santhosh Kumar Grandai.
1 Chapter 15 Probabilistic Reasoning over Time. 2 Outline Time and UncertaintyTime and Uncertainty Inference: Filtering, Prediction, SmoothingInference:
Bayesian networks and their application in circuit reliability estimation Erin Taylor.
Wei Sun and KC Chang George Mason University March 2008 Convergence Study of Message Passing In Arbitrary Continuous Bayesian.
Chapter 7 – PERT, CPM and Critical Chain Operations Management by R. Dan Reid & Nada R. Sanders 4th Edition © Wiley 2010.
Advances in Bayesian Learning Learning and Inference in Bayesian Networks Irina Rish IBM T.J.Watson Research Center
1 CMSC 671 Fall 2001 Class #20 – Thursday, November 8.
Probability. Probability Probability is fundamental to scientific inference Probability is fundamental to scientific inference Deterministic vs. Probabilistic.
Introduction on Graphic Models
COV outline Case study 1: Model with expert judgment (limited data): Perceived probability of explosive magmatic eruption of La Soufriere, Guadeloupe,
Bayesian Brain Probabilistic Approaches to Neural Coding 1.1 A Probability Primer Bayesian Brain Probabilistic Approaches to Neural Coding 1.1 A Probability.
Bayes for Beginners Anne-Catherine Huys M. Berk Mirza Methods for Dummies 20 th January 2016.
Outline Historical note about Bayes’ rule Bayesian updating for probability density functions –Salary offer estimate Coin trials example Reading material:
Chapter 12. Probability Reasoning Fall 2013 Comp3710 Artificial Intelligence Computing Science Thompson Rivers University.
Lecture on Bayesian Belief Networks (Basics)
Chapter 4 Probability.
Programming for Geographical Information Analysis: Advanced Skills
ICS 280 Learning in Graphical Models
Bayes for Beginners Stephanie Azzopardi & Hrvoje Stojic
Irina Rish IBM T.J.Watson Research Center
Markov ó Kalman Filter Localization
Class #21 – Monday, November 10
Bayesian Reasoning Chapter 13 Thomas Bayes,
Chapter14-cont..
Bayes for Beginners Luca Chech and Jolanda Malamud
CS639: Data Management for Data Science
Chapter 14 February 26, 2004.
Presentation transcript:

Copyright © 2009 PMI RiskSIGNovember 5-6, 2009 RiskSIG - Advancing the State of the Art A collaboration of the PMI, Rome Italy Chapter and the RiskSIG “Project Risk Management – An International Perspective”

Slide 2November 5-6, 2009 Copyright © 2009 PMI RiskSIG Bayesian Networks: A Novel Approach For Modelling Uncertainty in Projects By: Vahid Khodakarami

Slide 3November 5-6, 2009 Copyright © 2009 PMI RiskSIG Outline: u What is missing in current PRM practice? u Bayesian Networks u Application of BNs in PRM u Models u Case study

Slide 4November 5-6, 2009 Copyright © 2009 PMI RiskSIG Conceptual steps in PRMP u Risk Identification âQualitative Analysis u Risk Analysis (Risk Measurement) âQuantitative Analysis u Risk Response (Mitigation)

Slide 5November 5-6, 2009 Copyright © 2009 PMI RiskSIG Project Scheduling Under uncertainty u (CPM) u PERT u Simulation u Critical chain

Slide 6November 5-6, 2009 Copyright © 2009 PMI RiskSIG What is missing? u Causality in project uncertainty u Estimation and Subjectivity u Unknown Risks (Common cause factors) u Trade-off between time, cost and performance u Dynamic Learning

Slide 7November 5-6, 2009 Copyright © 2009 PMI RiskSIG Bayesian Networks (BNs) u Graphical model âNodes (variables) âArcs (causality) u Probabilistic (Bayesian) inference

Slide 8November 5-6, 2009 Copyright © 2009 PMI RiskSIG Bayesian vs. Frequentist FrequentistBayesian VariablesRandomUncertain Probability Physical Property (Data) Degree of belief (Subjective) InferenceConfidence interval Bayes’ Theorem only feasible method for many practical problems

Slide 9November 5-6, 2009 Copyright © 2009 PMI RiskSIG Bayes’ Theorem u ‘A’ represents hypothesis and ‘B’ represents evidence. u P(A) is called ‘prior distribution’. u P(B/A) is called ’Likelihood function’. u P(A/B) is called ’Posterior distribution’.

Slide 10November 5-6, 2009 Copyright © 2009 PMI RiskSIG Constructing BN High 0.7 Low 0.3 On time 0.95 Late 0.05 Prior Probability Sub-contract On time Late Staff Experience High Low High Low No Delay Yes Conditional Probability

Slide 11November 5-6, 2009 Copyright © 2009 PMI RiskSIG Inference in BN (cause to effect) With no other information P(Delay)= Knowing the sub-contract is late P(Delay)=0.50.7

Slide 12November 5-6, 2009 Copyright © 2009 PMI RiskSIG Backward Propagation (effect to cause) Prior probability with no data (0.7,0.3) Posterior (learnt) probability (0.28,0.72)

Slide 13November 5-6, 2009 Copyright © 2009 PMI RiskSIG BNs Advantages u Rigorous method to make formal use of subjective data u Explicitly quantify uncertainty u Make predictions with incomplete data u Reason from effect to cause as well as from cause to effect u Update previous beliefs in the light of new data (learning) u Complex sensitivity analysis

Slide 14November 5-6, 2009 Copyright © 2009 PMI RiskSIG BNs Applications u Industrial âProcessor Fault Diagnosis - by Intel âAuxiliary Turbine Diagnosis - by GE âDiagnosis of space shuttle propulsion systems - by NASA/Rockwell âSituation assessment for nuclear power plant – NRC u Medical Diagnosis âInternal Medicine âPathology diagnosis - âBreast Cancer Manager u Commercial âSoftware troubleshooting and advice – MS-Office âFinancial Market Analysis âInformation Retrieval âSoftware Defect detection u Military âAutomatic Target Recognition – MITRE âAutonomous control of unmanned underwater vehicle - Lockheed Martin

Slide 15November 5-6, 2009 Copyright © 2009 PMI RiskSIG Bayesian CPM CPM Calculation

Slide 16November 5-6, 2009 Copyright © 2009 PMI RiskSIG BCPM Example

Slide 17November 5-6, 2009 Copyright © 2009 PMI RiskSIG Activity Duration

Slide 18November 5-6, 2009 Copyright © 2009 PMI RiskSIG Trade off

Slide 19November 5-6, 2009 Copyright © 2009 PMI RiskSIG Trade off (Prior vs. required resources )

Slide 20November 5-6, 2009 Copyright © 2009 PMI RiskSIG Known Risk

Slide 21November 5-6, 2009 Copyright © 2009 PMI RiskSIG Known Risk (Control)

Slide 22November 5-6, 2009 Copyright © 2009 PMI RiskSIG Known Risk (Impact)

Slide 23November 5-6, 2009 Copyright © 2009 PMI RiskSIG Known Risk (Response)

Slide 24November 5-6, 2009 Copyright © 2009 PMI RiskSIG Unknown Factors

Slide 25November 5-6, 2009 Copyright © 2009 PMI RiskSIG Unknown Factors (Learning)

Slide 26November 5-6, 2009 Copyright © 2009 PMI RiskSIG Learnt distribution

Slide 27November 5-6, 2009 Copyright © 2009 PMI RiskSIG Total Duration

Slide 28November 5-6, 2009 Copyright © 2009 PMI RiskSIG Case Study (construction Project)

Slide 29November 5-6, 2009 Copyright © 2009 PMI RiskSIG Case Study (Bayesian CPM)

Slide 30November 5-6, 2009 Copyright © 2009 PMI RiskSIG Case Study (predictive)

Slide 31November 5-6, 2009 Copyright © 2009 PMI RiskSIG Case Study (diagnostic)

Slide 32November 5-6, 2009 Copyright © 2009 PMI RiskSIG Case Study (learning)

Slide 33November 5-6, 2009 Copyright © 2009 PMI RiskSIG Summary u Current practice in modelling risk in project time management has serious limitations u BNs are particularly suitable for modelling uncertainty in project u The proposed models provide a new generation of project risk assessment tools that are better informed and hence, more valid

Slide 34November 5-6, 2009 Copyright © 2009 PMI RiskSIG Questions? Thank you for your attention