Uncertainty Analysis Training

 Uncertainty Analysis Training

Uncertainty Analysis Training Course Description

The uncertainty analysis training course covers both theoretical and practical aspects of uncertainty measurements.

The uncertainty analysis training course will give you a step-by-step training and sufficient knowledge (a combination of information, skills, and tools) about uncertainty estimation that you can perform such assessment in your organization. Once you learn the fundamental concepts of uncertainty analysis, you can tune them based on your needs until they fit your work. The training comes with studying multiple real global projects along with statistical concepts. In addition, you will also exercise some theoretical cases during the course.

During the uncertainty analysis training course, you will learn how to apply risk and uncertainty analysis into decision making and project management. You will learn about statistics, probabilities, and estimation rules.  Such analysis is the basis of commercial decisions, as it demonstrates the possibility of the risk of success and failure of the decision and strategy that has been or about to be made. The uncertainty assessment also impacts the decision in quality control, operation of industrial plants, marketing, and even research and development.

In the uncertainty analysis training course, we will teach you the importance of traceability and repeatability of measurements in acceptability of the results.  We also will elaborate what is calibration and why it is essential to generate reliable readings.

The uncertainty analysis training course will teach you the methods required to perform measurement uncertainties and then apply these methods and statistical understanding to interpret the results.  The course consists of lectures, short exercises, and hands-on applications.

Audience

The uncertainty analysis training is a 2-day course designed for:

• Technicians and engineers
• Technical staff working in industries and calibration and test laboratories
• Auditors and personnel who do measurement traceability
• Individuals involved in both calibration and manufacturing processes

Training Objectives

Upon the completion of the uncertainty analysis training course, the attendees are able to:

• Discuss and understand the principals of measurement uncertainty calculations
• Perform practical measurements
• Understand the importance of uncertainty analysis
• Explain various statistical approaches
• Understand measurement functions
• Perform methods of evaluation uncertainty
• Explain sensitivity coefficients
• Determine effective degree of freedom
• Measure uncertainty and risk in different decision-making situations
• Understand the terminology associated with uncertainty analysis and measurement
• Understand the underlying mathematics required
• Perform basic techniques used in uncertainty analysis

Course Outline

Below are the topics will be covered in the uncertainty analysis training course. The topics can be tailored based on your organization’s needs.

Overview

• Uncertainty definition
• General metrological terms

Introduction to Cost Risk and Uncertainty

• Cost of risk and uncertainty
• Schedule risk and uncertainty
• Program risk management

Cost of Risk And Uncertainty

• Importance of cost risk and uncertainty
• Definitions
  • Cost Uncertainty
  • Cost Risk
  • Cost Uncertainty Analysis
  • Cost Risk Analysis
• Monte Carlo simulation
  • Point estimate
  • Correlation and its importance in simulation
  • Analysis of simulated estimate (PDF, CDF)

Schedule Risk Analysis

• Network-based schedules
• WBS-based schedules
• Schedule uncertainty
• Modeling uncertainty in schedules

Why Initial Cost Estimates are Almost Always Too Low

• Cost growth and its causes
• How cost growth impacts the analysis
• Cost growth vs. cost uncertainty
• Estimation of cost growth vs. estimation of cost uncertainty
• Anticipating cost growth and minimizing its impact

Probability Of Cost Analysis

• Importance of studying probability of cost estimation
• Information required to do such probability distribution
• Central Tendency measurement
  • Mean
  • Median
  • Mode
  • Difference between these three
• Variance measurement
  • Variance, standard deviation, coefficient of variation
  • How they are computed
• Probability distribution
  • Normal
  • Lognormal
  • Triangular
  • Uniform
  • Discrete
• Probability density functions and cumulative distribution functions
• Computing probabilities and percentiles from a cost probability distribution
  • Normal
  • Lognormal
• Some examples of cost probability distributions

Monte Carlo Simulation (OPTIONAL)

• What is it and how is it done?
• @RISK software tutorial
• Defining distribution
• Adding outputs
• The correlation matrix
• Settings
• Graphs

Understanding the Nature of CER and Cost Driver Uncertainty

• Uncertainty and cost estimation
  • Cost driver uncertainty
  • Cost estimating relationship (CER) uncertainty
  • The combined effect of both types of uncertainty on the cost estimate
  • Programmatic uncertainty
  • Technical uncertainty
  • Requirements uncertainty
• How to establish the uncertainty of CERs or other cost models
  • Regression results: standard error; standard percent error; prediction error
  • Subjective probability assessment for non-regression cost models
  • Distinguish between random and deterministic
• How to establish the uncertainty of cost drivers?
  • Data analysis
  • Ask informed people
  • Subjective probability analysis
• Briefly on subjective probability assessment
• Prediction intervals
  • Prediction intervals vs. confidence intervals
  • Employ prediction intervals to get a more accurate overall uncertainty
  • Prediction intervals limitations on non-OLS CERs
• An example of Monte Carlo simulation

The Correlation Effect

• Importance of developing correlation
• What do we mean by correlation anyway?
• The advantages of developing correlation in an assessment
• Different types of correlation
  • Pearson correlation
  • Spearman (rank-order) correlation
  • Causal correlation
  • Functional correlation
• How to model correlation in @RISK Monte Carlo simulation
• The correlation matrix
• Best practices
  • n2/2 problem
  • Book method
  • Anderson method
• How to test the consistency of a correlation matrix
  • Positive definite criteria
  • Eigenvalue method
• How to fix an inconsistent correlation matrix
• Example to practice

Schedule Risk Analysis

• What is schedule risk analysis
• What is Joint Cost/Schedule confidence levels
• Network-Based Schedules
  • How they are constructed
  • Critical path
  • Random vs. deterministic
  • How to model it in @RISK
• WBS-Based Schedules
  • How they are constructed
  • Random vs. deterministic
  • The WBS-Based schedule risk analysis process
  • How to model it in @RISK
• Joint Cost/Schedule Confidence Levels (JCLs)
• Schedule risk analysis vs. phasing

Phasing the Cost Estimate

• What does it mean?
• Phase estimates methods
  • Rayleigh distribution for pre-milestone A estimates
  • WBS-based phasing for existing acquisition programs
  • Schedule Estimating Relationships
  • Outlay profile method
  • Other phasing algorithms
• How to convert a cost profile to a budget profile
• Phase a pth percentile estimate
  • Using the program schedule
  • Using the expected Rayleigh distribution
• How to convert an expenditure profile to a budget profile
• A pth percentile estimate allocation back to the WBS elements

Project/Case Study