Scribd
Upload
11 views

Simulation

The document discusses simulation and provides examples of its use. It defines simulation as attempting to duplicate real systems through mathematical models. Key points include: 1. Simulation allows studying systems that cannot be solved through conventional models by including real-world complications. 2. Common mistakes in simulation include lacking clear objectives, poor data collection, and failing to question results. 3. Monte Carlo simulation uses random sampling to model probabilistic systems. It sets probability distributions and runs random trials. 4. An example simulates unloading barges over 15 days, finding the average delays, arrivals, and unloads per day to inform staffing decisions.

Uploaded by

Syama Aliyah Bentor
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
11 views

Simulation

The document discusses simulation and provides examples of its use. It defines simulation as attempting to duplicate real systems through mathematical models. Key points include: 1. Simulation allows studying systems that cannot be solved through conventional models by including real-world complications. 2. Common mistakes in simulation include lacking clear objectives, poor data collection, and failing to question results. 3. Monte Carlo simulation uses random sampling to model probabilistic systems. It sets probability distributions and runs random trials. 4. An example simulates unloading barges over 15 days, finding the average delays, arrivals, and unloads per day to inform staffing decisions.

Uploaded by

Syama Aliyah Bentor
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Simulation Top 10 List of Simulation NO-NOs

 Performing a simulation without a clear definition of the


Simulation objectives;
 is the attempt to duplicate the features, appearance and  Believing that the model itself can compensate for poor
characteristics of a real system. data collection;
 it involves the building of a mathematical model that  Lacking an understanding of statistical processes;
comes as close as possible to representing the reality of  Failing to do order profiling;
the system.  Ignoring the effects of randomness;
 The idea is to imitate a real world situation  Incorporating randomness inappropriately;
mathematically, then to study its properties and  Failing to consider down time;
operating characteristics, and finally to draw  Making illogical assumptions;
conclusions and make action decisions based on the  Failing to question the results;
results of the simulation.  Failing to recognize that simulation is a study tool

 The problems tackled by simulation may range from The Simulation Process
very simple to extremely complex. Although very small 1. Define the problem.
simulations may be conducted by hand, effective use of 2. Identify and introduce important variables.
this technique requires some automated means of 3. Construct the simulation model.
calculation with the aid of a computer software. 4. Specify the values of the variables to be tested
5. Conduct simulation.
 While simulation is one of the older Management 6. Examine the results.
Science techniques, it was not until the introduction of 7. Select the best course of action.
computers that it became a practical means of solving
large-scale problems. Monte Carlo Simulation
 When a system contains elements that exhibit chance in
Simulation their behavior, the Monte Carlo method of simulation
 can handle queuing problems in which arrivals are not may be applied.
Poisson distributed, service time are not exponentially  The basis of Monte Carlo simulation is experimentation
distributed or constant, or the FIFO priority system is on the chance (or probabilistic) elements of the problem
inappropriate. through random sampling

Advantages of Simulation Monte Carlo Steps.


 It is relatively straightforward and flexible; 1. Set up a probability distribution.
 It can be used to analyze large and complex real world 2. Build cumulative probability distribution for each of the
situations that cannot be solved by conventional variables.
Management Science models. 3. Establish an interval of random numbers for each
 Simulation allows for the inclusion of real world variable.
complications that most Management Science models 4. Generate random numbers.
require exponential or Poisson distributions. Some 5. Simulate a series of trials.
models require normality and standard deviations, the
simulation does not. Step 1
 “Time compression” is possible with simulation.
 Allows “what-if” types of questions;
 Do not interfere with the real world system. With
simulation, experiments are done with the model, now
on the system itself; and
 Allows us to study interactive effect of individual
components or variables in order to determine which
ones are important.

Disadvantages of Simulation
 Good simulation models can be very expensive; Step 2
 Simulation may not generate optimal solutions to
problems as like Linear Programming;
 Each simulation model is unique. Its solutions and
inferences are not usually transferable to other
problems or situations.
Step 3

Step 4

Step 5

 3 barges could have been unloaded on day 2. But


because there were no arrivals and no backlog existed,
Of course, it would be risky to draw any hard and fast zero unloading took place. The same explanation goes
conclusions regarding the operation of a firm from only a for day 7
short simulation. It is also unlikely that anyone would actually  4 barges could have been unloaded on day 14, but since
want to go through the effort of simulation such a simple only 3 were in queue, the number of unloaded is only 3.
problem containing only one variable.
The superintendent would most likely be interested in at
Simulating by hand does, however, demonstrate the least 3 useful and important pieces of information:
important principles involved and may be useful in small- 1. The average number of barges delayed to the next day
scale studies. As one might expect, the computer can be a 20 delays per 15 days = 1.33 barges delayed per day
very helpful tool in carrying out the tedious work in larger 2. The average number of nightly arrivals
simulation undertakings. 41 arrivals per 15 days = 2.73 arrivals per night
3. The average number of barges unloaded each day
Another Example 39 unloading per 15 days – 2.60 unloading per day.

When the above data are analyzed in the context of delay


costs, idle labor costs, and the cost of hiring extra unloading
staff, it will be possible for the superintendent and the
controller to make a better staffing decision. They may even
chose to re-simulate the process assuming different
unloading rates that would correspond to the increase in
crew size.

While simulation does not guarantee an optimal solution to


problems, such as the one above, it can be helpful in
recreating a process and identifying good decision
alternatives.

You might also like