Risk & Optimization using RISKOptimizer

This one day training course is an addition to the existing @Risk course.  It will provide attendees with an understanding to effectively use RISKOptimizer to solve complex problems that involve uncertainty.  RISKOptimizer uses Genetic Algorithms to work with a large numbers of variables and rules to find optimal solutions for a given problem.  These problems are often every day issues dealing with production scheduling, inventory utilization, and delivery route determination – any situation where you have a process that can be optimized, but is very complex; RISKOptimizer can help.  The course starts by providing attendees with a simple yet complete understanding of how the optimization engine of RISKOptimizer works, and the benefits it can provide an organization.  Once the optimization engine is understood, the course then examines how this powerful functionality is integrated into Microsoft Excel for an easy to use interface.  Several exercises will be completed during the class to ensure that students have a good understanding of the product and how it can benefit their organization.

Day 1

Introduction

  1. Introduction to RISKOptimizer
    1. Concept
      1. Optimization Using Genetic Algorithms
      2. Mechanics
      3. Addition of uncertainty
  2. RISKOptimizer Interface
    1. Configuring Optimization Settings
      1. Variables
      2. Constraints
      3. Goal
    2. Configuring Probability Distributions
      1. Common Distribution Functions
  3. Creating a Simple Model
    1. Optimization Model
    2. Adding Uncertainty
    3. Running a Model
    4. The Results
    5. A simple exercise

Optimization

  1. Basics
    1. Inputs
      1. Population of values that are adjusted to find the optimal goal
      2. Values that may be subject to constraints
        1. Soft
        2. Hard
        3. Warnings about constraints
    2. Functions
      1. May include data from other sources
      2. This is your calculation that will in a output goal
      3. Excel formulas that may be very complex
    3. Outputs
    4. Exercise
  2. Hill Climbing Approach
    1. Local Solution verse Global Solution
  3. Genetic Algorithms (GA)
    1. Biological Terms
      1. Population
      2. Chromosome
      3. Gene
      4. Mutation
      5. Crossover
      6. Fitness
    2. Example of GA
  1. Exercise

Adding Uncertainty to Optimization with RISKOptimizer

  1. Why does addressing Risk make the model better?
  2. How does simulation work with optimization?
  3. Identifying areas of risk
    1. Variables
      1. Independent and Dependent
  4. Assigning a Probability Distribution
    1. Fitting a distribution to a set of data
      1. Often with models you don’t know the best distribution to use; in fact all you have is a list of data points.  With RISKOptimizer you can analyze that data set and have the application provide your model with the best distribution.  This is especially helpful if your input data sets change frequently.
  5. Fitting uncertainty into your optimization model
  6. Exercise

Advance Topics

  1. Using macros
  2. Scheduling Models
  3. Development Kits

Review and Conclusion

 

» Palisade Corporation
» Seminar Schedule