Print

Quantitative Risk Assessment Methods: Part 1 - SIP

Register

Sign up for this course now, by clicking here.

Course Description

Quantitative Risk Assessment Methods: Part 1 teaches students the basics of building and understanding quantitative risk assessment models. This two-and-a-half day course will cover basic modeling concepts, including both deterministic and probabilistic modeling approaches. Students will be taught how to build risk assessment models using Excel. This course will also introduce students to one of the more commonly-used commercial software packages (@RISK™)

Students will learn both why and how to include variability and uncertainty in models. This requires a review of basic probability theory and the concept of Monte Carlo simulation. They will then apply these concepts by describe model variables using probability distributions and generating probabilistic risk estimates (using @RISK™)as an add-in to Microsoft Excel).

Learning by example, students will be given exercises involving elements of real world risk assessments that are being used in current policy and risk management.

The course is conducted in a computer teaching laboratory with two instructors. Lectures will describe various techniques. Students then work individually and in groups to solidify their understanding of the lecture materials and to build quantitative modeling skills.

Students should have basic knowledge of probability and statistics and intermediate level skills in using Microsoft Excel 2003. We also strongly recommend Food Safety Risk Assessment as a prerequisite to this course.

Resources

Excel: There are web-based resources that provide introductory Excel 2003 training. Many such courses are available - some at no cost - like the one found at www.videoprofessor.com.

Basic Statistics: The quantitative methods courses do not require in-depth knowledge of statistics, but an understanding of basic terminology is necessary. There are web-based resources that provide information about basic theory in probability and statistics. Some examples include http://www.robertniles.com/stats and http://www.statsoft.com/textbook.

Overview of Topics

Introduction to modeling

What is a model?

  • Different types of models

Why do we build quantitative models?

  • Candidate objectives of a modeling exercise
  • What are the alternatives to quantitative models?
  • Some examples of their use in real decisions

Key considerations in the design of risk assessment models

  • accuracy, complexity, transparency and other issues
  • risk assessment as a decision-support product

Deterministic modeling

  • What is a deterministic model?
  • Exercise building a simple deterministic model in Excel
  • More complex deterministic models
  • Sensitivity and scenario analysis with a deterministic model
  • Case study: analysis and critical review of a deterministic model

Probabilistic modeling

  • Considering variability and uncertainty
  • What is a probabilistic model?
  • Why do we need probabilistic models?
  • What are the alternatives?

Expressing variability and uncertainty using probability distributions

Definitions

  • Discrete versus continuous distributions
  • Some common distributions (uniform, normal, etc.)
  • Exercise: describing everyday phenomena as distributions

Interpreting graphical representations of probability

  • Probability density and cumulative distributions

Review of Statistical Measures

  • Mean, Median, Mode, Percentiles

Introduction to Monte Carlo Simulation

  • How does it work?
  • How does add-in software work with Excel?
  • Building a simple simulation model using @RISK™
  • Exploring probability distributions using @RISK™
  • Understanding @RISK™ Output
  • Exercises
    • Microbial risk assessment example
    • Chemical risk assessment example
    • Comparing deterministic and probabilistic models of the same problem

A brief demonstration of other modeling environments

  • Open source software (R, Octave, WinBugs, etc.)
  • Other Commercial Software (Crystal Ball, Analytica, Matlab, etc.)
  • Pros and Cons in choosing a modeling environment

Summary: Designing Risk Assessments to Support Decision-Making

Learning Objectives

After completing this course, students will:

  • Understand why models are useful
  • Understand important tradeoffs in the design of models
  • Understand the differences between deterministic and stochastic models
  • Gain a strong foundation in basic probability theory and probability distributions
  • Be able to build basic probabilistic models using Excel and @RISK™
  • Understand some pros and cons of other model-building environments