SIMULATION APPLICATION The concept of how and when to use simulation evolves with time.

The early years In the late 1950s, simulation was a very expensive and specialized tool that was generally used only by large corporations that required substantial capital investments. Typical simulation users were found in aerospace and steel corporations. They would form groups of people mostly people with Ph.D.s,who would develop and solve the complex simulation models using specific language such as FORTRAN. The models would be run on large and expensive mainframes charging from $600 to $1000 per hour. The formative years During 1970s computer becoming faster and cheaper and the concept of simulation was being discovered by other industries. When there was a disaster, simulation would be considered. It became the tool of choice for many companies especially in the automotive and heavy industries. During this time, simulation also found a home in academia as a standard part of industrial engineering and operation research curricula and started to broadening the number and type of students and researchers exposed to its potential. The recent past During the late 1980s, simulation began to establish its real roots in business in line with the introduction of the personal computer and animation. Instead of using the simulation to analyze failed system, some was requesting simulation before production was to begin. However, simulation was still not in widespread use and was rarely used by smaller firms. The present During 1990s, simulation began to mature. Many smaller firms embraced the tool and it began to see use at the very early stages of projects where it could have the greatest impact. Better animation, greater ease of use, faster computers, easy integration with other packages and the emergence of simulators have all helped simulation become a standard tool in many companies. Besides, people are more interested in conducting researchs on simulation. As a result, the application of simulation has widely spread.

In manufacturing, simulations are used as a reliable tools to measure system performance. System performance includes measuring utilization of resource, labor and

machines,effectiveness of scheduling system,effectiveness of control systems and queing at workplace. In clinical healthcare, B.A Peters et. al conducted a study on a clinic with a financial constraints. They describes the creation of a low-cost, generic, discrete-event simulation model populated by a workflow observation Excel spreadsheet that can be completed by clinic staff themselves, thus customizing the simulation model for their own purposes. This model focuses on childhood immunization delivery services where the intent is to develop a tool flexible enough to serve other health services delivery needs as well. While in 2002, Bart W.M. Heesbeen et.al conducted a research on flight simulation of future autonomous aircraft opeartions. Research flight simulation is applied when the performance and perception of human pilots is a key measure of the overall assessment. The research conducted give an overview of the research simulation set-up of the National Aerospace Laboratory (NLR), Amsterdam, the Netherlands, which is used for the human-in-the-loop evaluation of future operational concepts. Special attention is given to the research topic of Airborne Separation Assurance where often referred to as Free Flight.

The future In recent years, the rate of change in simulation has accelerated and there is every reason to believe that it will continue its rapid growth and cross the bridges to mainstream acceptance.

SIMULATION APPROACHES There are many options for running a simulation that evolve from time to time. By hand In the early days, people did do simulations by hand. Georges Louis leclerc conducted an experiment to estimate the value of . A needle of lenght l was tossed onto a painted table with parallel lines spaced d apart.The number of times the needle would tossed have to be decided and the counter was set to 0.Then the needle was tossed randomly. If it crossed one of the lines, the counter would be added by 1. It it does not crossed the line, leave the counter alone.Then the proportion of times the needle crossed a line was computed and resulted in . Programming in General-purpose languages. This approach was highly customizable and flexible and appeared in the 1950s and 1960s. People began writting computer programs in general-purpose procedural languages like FORTRAN to do simulations of more complicated system. Simulation Languages Specific-purpose simulation languages like GPSS, Simscript, SLAM and SIMAN appeared to provide a much better framework for the kinds of simulations many people do. All you need to do is to invest quite a bit of time to learn about their features and how to use them effectively.

High-Level Simulators They typically operate by intuitive graphical user interfaces, menus, and dialogs. You select from available simulation-modeling constructs, connect them, and run the model along with a dynamic graphical animation of system components as they move around and change.

DESIGNING SIMULATION EXPERIMENT

A simulation experiment is a test or a series of tests in which meaningful changes are made to the input variables of a simulation model in order to observe and identify the reasons for changes in the performance measures. The number of experiments in a simulation study is greater than or equal to the number of questions being asked about the model. The following steps illustrate the process of designing a simulation experiment. Step 1 : Select the appropriate experimental design. Select a performance measure, a few input variables that are likely to influence it, and the levels of each input variable. When the number of possible configurations (product of the number of input variables and the levels of each input variable) is large and the simulation model is complex, common second-order design classes including central composite, BoxBehnken, and fullfactorial should be considered. Document the experimental design. Step 2 : Establish experimental conditions for runs. Address the question of obtaining accurate information and the most information from each run. Determine if the system is stationary (performance measure does not change over time) or non-stationary (performance measure changes over time). Generally, in stationary systems, steady-state behavior of the response variable is of interest. Ascertain whether a terminating or a non-terminating simulation run is appropriate. Select the run length. Select appropriate starting conditions. If required, select the length of the warm-up period. Decide the number of independent runs - each run uses a different random number stream and the same starting conditions -by considering output data sample size. Sample size must be large enough (at least 3-5 runs for each configuration) to provide the required confidence in the performance measure estimates. Alternately, use common random numbers to compare alternative configurations by using a separate random number stream for each sampling process in a configuration. Identify output data most likely to be correlated. Step 3 : Perform simulation runs.

References: Kelton, W.D.,Sadowski, R.P.,& Sturrock, D.T (2010). Simulation with Areana (5th ed). New York : McGraw-Hill B.A Peters, J.S.Smith, D.J Modeiros and M.W. Rohrer eds (2001). A discrete-event simulation application for clinics serving the poor. Mario.S.V.Valenti Clariu,Rob C.J. Ruigrok, Bart W.M Heesbeen, Jaap Groeneweg (2002). Research flight simulation og future autonomous aircraft operations.