RISK
ANALYSIS
SIMULATION
PROBABILITY & STATISTICS FUNDAMENTALS
Probability - Probability is a measure of how likely a Confidence Interval - Confidence intervals refers an
value or event is to occur interval that provides a range where the exact or true
probability value of parameters may lie
Probability Density Function (PDF) - A function Frequency Distribution or Histogram - A graphic
representing the probability distribution of a (plot) summarizing the frequency of the values
continuous random variable observed or measured from a population
Probability Mass Function (PMF) - A function
representing the probability distribution for a discrete Monte Carlo Analysis (MCA) or Monte Carlo
random variable Simulation - A technique for characterizing the
uncertainty and variability in risk estimates by
repeatedly sampling the probability distributions of the
Random Variable - A variable that may assume any
value from a set of values according to chance risk equation inputs and using these inputs to calculate
a range of risk values
Cumulative Distribution Function (CDF) - Obtained
by integrating the PDF, gives the cumulative
probability of occurrence for a random independent
variable
PROBABILITY DISTRIBUTION TYPES
DISCRETE
Uniform, Binomial, Negative
Binomial, Hypergeometric, Poisson
DISTRIBUTION
CONTINUOUS
Normal, Exponential, Gamma,
Weibull, Rayleigh, Lognormal
Refers to your probability and statistics
notes for detailed explanations
SIMULATION
Simulation is a process of understanding behavior of a system or evaluating
the various strategies for the operation of the system.
Simulation calculates multiple scenarios of a model by repeatedly sampling
values from the probability distributions for the uncertain variables.
Simulation involves the generation of the artificial history of the system and
observation of the artificial history to draw inferences concerning the
operating characteristics of the real system that is presented.
WHY
SIMULATION
1
When it is too costly to do physical studies on the
system itself (e.g., trying alternative layout of a
factory, building new facility)
2
The corresponding analytic models are too
complicated to study (e.g. a
queuing/transportation network)
3 When uncertainty, non-stationarity dominates
the output results.
SIMULATION TOOLS
@Risk Matlab Minitab
Ms. Crystal
SPSS
Excel Ball
RANDOM NUMBERS & PROBABILITY
DISTRIBUTION IN SIMULATION
It is a numerical description of the outcome of some experiment. It can be
generated from a probability distribution.
In simulation terminology, a random number is one that is uniformly
distributed in between 0 and 1.
Recall from statistics that uniform probability distribution characterizes a
random variable for which all outcomes between a minimum value a and a
maximum value b are equally likely
RANDOM NUMBERS & PROBABILITY
DISTRIBUTION IN SIMULATION
DISCRETE Rarely used in engineering problems
CONTINUOUS
Inverse transform method is used in case if there is no direct function to generate random numbers.
Inverse transform method uses the following steps:
1. generate a random number from the Uniform distribution: u=Uniform(0,1),
2. Calculate inverse cumulative distribution function (ICDF). In Excel, the function name to
calculate the ICDF for Normal distribution NORMINV and for LogNormal distribution is
Lognormal (LOGINV). RAND function in excel can also be used to generate random numbers
from the Uniform distribution, and apply the built-in functions to calculate the ICDF.
MONTE-CARLO
SIMULATION
MONTE-CARLO SIMULATION
Monte-Carlo Simulation uses the
following steps for assessing
parameters uncertainty:
• Select a distribution to describe
possible values of a parameter.
• Generate data from this
distribution.
• Use the generated data as probable
values of the parameters in the
model to produce output.
SIMULATION USING
MS. EXCEL
Basic Excel Skills
To facilitate the understanding of simulation using Excel, it is necessary to be familiar with the following features of Excel
• Copying formula and cell references
• Function: It is used in performing special calculations in the Spread sheet cells. Some of the more common functions
that are usually used in simulation models are including:
✓ MlN (range) - finds the smallest value in a range of cells.
✓ MAX (range) - finds the largest value in a range of cells
✓ SUM (range) - finds the sum of values in a range of cells
✓ AVERAGE (range) - finds the average of the values in a range of cells
✓ STDEV(range) -finds the standard deviation for a sample in a range
of cells
✓ AND (condition 1, condition 2,) -a logical function that returns TRUE if all conditions are true, and FALSE
if not.
✓ OR (condition1, conditiott2 . . .) - a logical function that returns
TRUE if any condition is true, and FALSE if not.
✓ IF (condition, value if true, value if false) -a logical function that
returns one value if the condition is true and another if the condition is false.
✓ VLOOKIUP (value, table range column number) -looks up a value in a table.
• Charts and graphs
• Excel has many other functions for statistical, financial and other applications
Some commonly used distributions
Commonly used distributions
Example
Profit is calculated using equation below
Profit = Revenue – (Variable Cost + Fixed Cost)
Using the distributions properties below, find the expected value
(mean), median, minimum, and maximum values of profit. Use 100
simulations.
Factor Distribution Mean Standard
Deviation
Revenue Normal 5,000,000 50,000
Fixed Cost - 100,000
Variable Cost Normal 200,000 20,000
SIMULATION USING
CRYSTAL BALL
Installing Crystal Ball
Download Crystal Ball. If you have not already done so, download Crystal Ball
from:
https://www.oracle.com/middleware/technologies/crystalball/downloads.html
License Crystal Ball. ** Please note that Crystal Ball has adopted new licensing
technology. Here is your Username and Serial Number: **
Product: Crystal Ball
Username: ULI-CBWebinar
Serial Number: EF6534A2-46B74E42-468BBBFA-2D658801
Version: N/A
License for: single user Features: OPTIMIZATION WATERMARK CLASSROOM
License Type: Trial
Duration: 120 day(s)
Example
Profit is calculated using equation below
Profit = Revenue – Cost
Using the distributions properties below, find the expected value
(mean), median, minimum, and maximum values of profit. Use 10000
simulations.
Variable Distributi Mean Std Min Most Max
on Deviation Likely
Revenue Normal 500 50 - - -
Cost Triangular - - 200 300 400
Also perform following analysis
1) What is the likelihood of achieving a profit of 200?
2) What is the likelihood of achieving no profit?
3) What profit range is obtained with certainty of 80%.
4) Perform sensitivity
