## Steps (Algorithm)

* 1. Formulate the model.* This step is almost the same as that
for other operations research models. For a proper formulation, comprehensive
study should be made regarding components of the problem, objective,
composition of the organisation, etc.

Never add too much detail in a simulation model, it will consume a lot of computer time. Always keep your specific goal in mind.

*2. Design the experiment. * After building a **simulation model**,
simulation experiments must be designed. In this step, you must decide
the starting conditions of the model, parameter settings, time period
required for each run, total number of runs, etc.

Work out the details of the experimental procedures before running the model, it will considerably reduce the chances of making blunders. Think carefully about the model to save your time and money

*3. Develop the Computer Program:* Using a high-speed electronic
computer, simulation experiments can be performed. If the simulated
model has a very simple structure, you can use a standard programming
language, such as FORTRAN, PL/1, or ALGOL, to develop the computerized
version. On the other hand, for a complex structure, you can use simulation
languages like SIMSCRIPT or GPSS.

Never assume blindly that the entire simulated system is accurate & reliable, merely because each of the component parts seems accurate when considered in isolation.

## Monte-Carlo Simulation

The **Monte-Carlo simulation method** uses random numbers for generating
some data by which a problem can be solved. These random numbers are
helpful in creating a new set of hypothetical data for a problem whose
behaviour is known from past experience. The random numbers are generated
either on a computer or are picked up from a table. Most computers employ
what is known as pseudorandomness. This means that the numbers are generated
by a series of specific operations. Each number is generated by performing
these operations on the previous number. After picking a random number,
its value is compared with the cumulative probability distribution and
the value of process parameters is obtained.

The cumulative probability distribution is most important instrument in the use of Monte-Carlo methods.