SIMULATION: PRESCRIPTIVE

ANALYTICS
 Learning Objective:
At the end of the lesson, the student should be able to:
• Become familiar with Monte Carlo Simulation.
• Apply the 5 steps of Monte Carlo Simulation.
• Conduct Monte Carlo Simulation using Spreadsheets.
Prescriptive Analytics
Prescriptive analytics use optimization algorithms
and simulation to quantify the effect of different
possible actions of a decision-maker to make a more
informed decision.
The focus of this module is simulation,
particularly using Monte Carlo Simulation.
Simulation
A simulation is an alternative form of analysis. It is a model
that mimics the operation of an existing or proposed system,
providing evidence for decision-making by being able to test
different scenarios or process changes (Taylor, 2013).
It is the process of designing a model of a real system and
conducting experiments with this model for the purpose of
understanding the behavior (within the limits imposed by a
criterion or a set of criteria) for the operation of the system.
Simulation
Simulation is used since not all real-world problems can be
solved by applying a specific type of technique and then
performing the calculations. Also, some problem situations
are too complex to be represented by the concise techniques
previously taken up in this course.
Simulation (Anderson et al, 2016)
It is a method for learning about a real system by
experimenting with a model that represents the system.
The simulation model contains the mathematical expressions
and logical relationships that describe how to compute the value
of the outputs given the values of the inputs.
The inputs for a simulation model can be classified as
controllable or uncontrollable inputs. Uncontrollable inputs can
be either uncertain or known with certainty.
Simulation (Anderson et al, 2016)
Conceptual diagram of a simulation model
Types of Simulation
Analogue simulation is the most familiar type of simulation.
It replaces a physical system with an analogous physical
system that is easier to manipulate.
Examples:
▪ Experimentation in staffed space flight using physical
simulation that re-created the conditions of space.
▪ Conditions of weightlessness were simulated using rooms
filled with water.
▪ https://youtu.be/Ft3QR0Ie9YA
Types of Simulation
Computer mathematical simulation, the focus of this
module, uses a mathematical model that is analyzed by using
the computer, to replicate a system.
Examples:
▪ Used in corporate finance to model components of project
cash flow, which are impacted by uncertainty. The result is a
range of net present values (NPVs) that can be used by an
investor to estimate the probability that NPV will be greater
than zero.
Types of Simulation
Examples:
▪ It allows an analyst to determine the size of the portfolio a
client would need at retirement to support their desired
retirement lifestyle and other desired gifts and bequests.
With inflation rates, tax rates, and reinvestment rates
factored in, the result is a distribution of portfolio sizes with
the probabilities of supporting the client’s desired spending
needs.
The Monte Carlo Simulation: Understanding the Basics (investopedia.com)
Applications of Simulation
▪ New Product Development
The objective of this simulation is to determine the probability
that a new product will be profitable.
-A model is developed that relates profit (the output
measure) to various uncertain inputs such as demand, parts
cost, and labor cost.
-The only controllable input is whether to introduce the
product.
Applications of Simulation
▪ Airline Overbooking
The objective of this simulation is to determine the number of
reservations an airline should accept for a particular flight.
-A model is developed that relates profit for the flight to an
uncertain input, the number of passengers with a reservation
who show up and use their reservation, and a controllable
input, the number of reservations accepted for the flight.
Applications of Simulation
▪ Inventory Policy
The objective of this simulation is to choose an inventory
policy that will provide good customer service at a reasonable
cost.
-A model is developed that relates two output measures,
total inventory cost and the service level, to uncertain inputs,
such as product demand and delivery lead time from vendors,
and controllable inputs, such as the order quantity and the
reorder point.
Applications of Simulation
▪ Waiting Lines
The objective of this simulation is to determine the waiting
times for customers requesting service from a facility, such as
customers phoning a call center.
-A model is developed that relates customer waiting times to
uncertain inputs, such as customer arrivals and service times,
and a controllable input, such as the number of servers (such
as call center agents).
Monte Carlo Simulation
It was invented during World War II by John von Neumann
and Stanislaw Ulam to improve decision-making under certain
conditions. This technique involves assigning values to an
uncertain variable to achieve multiple results and then averaging
the results to obtain an estimate.
It finds its application in prediction and forecasting models in
business, supply chain, project management, finance, science,
engineering, particle physics, artificial intelligence, astronomy,
meteorology, sales forecasting, and stock pricing.
What is Monte Carlo Simulation? | IBM
Monte Carlo Simulation
The name Monte Carlo is appropriate because the basic
principle behind the process is the same as in the operation of a
gambling casino in Monaco. In Monaco, such devices as roulette
wheels, dice, and playing cards are used. These devices produce
numbered results at random from well-defined populations.
The same process is employed, in principle, in the Monte Carlo
process used in simulation models.
Monte Carlo Simulation
The Monte Carlo method is a stochastic (random sampling
of inputs) method to solve a statistical problem, and a simulation
is a virtual representation of a problem. The Monte Carlo
simulation combines the two to give us a powerful tool that
allows us to obtain a distribution (array) of results for any
statistical problem with numerous inputs sampled over and over
again.
The Monte Carlo Simulation: Understanding the Basics (investopedia.com)
Monte Carlo Simulation (Render et al, 2018)
The basic idea in Monte Carlo Simulation is to generate
values for the variables making up the model being studied.
Below are some examples of variables in real-world systems that
we might want to simulate:
1. Inventory demand on a daily or weekly basis
2. Lead time for inventory orders to arrive
3. Times to complete project activities
4. Number of employees absent from work each day.
Five Steps of Monte Carlo Simulation (Render et al, 2018)
1. Establishing a probability distribution for important input variables.
2. Building a cumulative probability distribution for each variable in Step 1.
3. Establishing an interval of random numbers of each variable.
4. Generating random numbers.
5. Simulating a series of trials.
The Use of Random Numbers
Random numbers are numbers equally likely to be chosen
from a large population of numbers.
In a random number table, the random numbers are derived
from some artificial process, like a computer program.
The Use of Random Numbers
Example:
Harry’s Auto Tire sells all types of tires, but a popular radial tire
accounts for a large portion of Harry’s overall sales. Recognizing that
inventory costs can be significant with this product, Harry wishes to
determine a policy for managing this inventory. To see what the demand
would look like over a period of time, he wishes to simulate the daily
demand for 10 days.
Example:
Step 1: Establishing probability distribution.
One common way to establish a probability distribution for a given
variable is to examine historical outcomes. The probability, or relative
frequency, for each possible outcome of a variable, is found by dividing
the frequency of observation by the total number of observations. The
daily demand for radial tires at Harry’s Auto Tire over the past 200 days
is shown in the next slide.
Example:
Step 1: Establishing probability distribution.
The frequencies are converted to a probability distribution, if we assume
that past demand rates will hold in the future, by dividing each demand by
the total demand.
Example:
Step 2: Building a Cumulative Probability Distribution for Each
Variable.
A cumulative probability is the probability that a variable will be less than
or equal to a particular value. The table shows the cumulative probability for
radial tires.
Example:
Step 2: Building a Cumulative Probability Distribution for Each
Variable.
The graph of the cumulative
probability will be used in Step 3
to help assign random numbers.
Example:
Step 3: Setting random number intervals.
In general, the cumulative
probability distribution and
graph obtained from Step 2 will
be used to set the interval of
random numbers for each level
of demand in a very simple
fashion. In the table, the
interval selected is very closely
related to the cumulative
probability on its left. The top
end is always equal to the
cumulative percentage.
Example:
Step 3: Setting random number intervals.
The length of each interval
on the right corresponds to
the probability of one of
each of the possible daily
demands. Hence, in
assigning a random number
to the daily demand for 3
tires, the range of 36 to 65
corresponds exactly to the
probability of that outcome.
Example:
Step 3: Setting random number intervals.
Daily demand for three radial
tires occurs 30% of the time.
One of the 30 random
numbers greater than 35 up
to and including 65 is
assigned to the event.
Example:
Step 4: Generating random
numbers.
Random numbers may be
generated either using a
calculator, a table of random
numbers, or using the
function”=Rand()” in Excel.
In the table of random
numbers, we can select from
anywhere in the table to use
in our simulation procedures
in Step 5.
Example:
Step 5: Simulating the
Experiment.
Beginning anywhere in the
table, we note the interval in
slide 25 into which each
number falls. For example, if
the random number chosen is
81 and the interval 66 to 85
represents a daily demand for
4 tires, we select a demand
for 4 tires.
Example:
Step 5: Simulating the Experiment.
We now illustrate the concept
further by simulating 10 days
of the demand for radial tires.
We select the random
numbers needed from the
table, starting in the upper
left-hand corner and
continuing down the first
column.
Example:
Step 5: Simulating the Experiment.
It is interesting to note that the average demand of 3.9 tires in the 10-day
simulation differs significantly from the expected daily demand, computed as
follows from the table in slide 24:

If this simulation were repeated hundreds or thousand times, it is much

more likely that the average simulated demand would be nearly the same as
the expected demand.
References
• Quantitative Analysis for Management 12th Edition by Render et al
• Introduction to Management Science by Taylor
• Best Excel Tutorial - Monte Carlo simulation in Excel (best-excel-
tutorial.com)
• Introduction to Monte Carlo simulation in Excel - Microsoft Support
• How To Add Monte Carlo Simulation to your Spreadsheet Models |
RiskAMP
• Creating a Monte Carlo Simulation Using Excel (investopedia.com)
• How to Create Monte Carlo Models and Forecasts Using Excel Data
Tables - ExcelUser.com
• Monte Carlo Simulation in Excel: A Practical Guide (vertex42.com)
• Introduction to Monte Carlo simulation in Excel - Microsoft Support