Whether designing a new system or modifying an existing one, decision-makers want to take the guesswork out of finding the best possible solution. Simulation makes it possible to study, analyze, and evaluate different alternative solutions.

Simulation is a duplication of reality. A simulation creates a representation of an actual system (either proposed or existing) in order to investigate its key performance measures. It is the art and science of translating a real-world system into a representative model. This model is then used and using the latter as a convenient test bed to evaluate the behavior of a real-world system. Simulation is particularly well-suited for recommending system design parameters.

Representation is the modeling approach used to capture essential characteristics and functions of the actual system under study. A real-world system can be represented in either a physical or computer model.

A physical model is a representation of the system, in scale or in function. For example, a physical model of an airplane is used in a wind tunnel to test the plane's aerodynamics. A computer simulation on the other hand, uses programming tools to capture the main logical and mathematical relationships that govern the functions of the actual system. Some simulations combine both types of models, such as a flight simulator used to train pilots. The simulator interior reproduces the physical characteristics of the cockpit, and a complex system of hydraulics simulates an airplanes physical movements. At the same time, a computer creates the simulated environments and the reacts to the pilot's inputs.

It is important to remember the different between simulation and animation. Animation can be a component of effective simulation; however, an animation is not the same as simulation. In fact, a computer program can be written to represent a certain material handling process and collect valid key performance statistics, without any animation. On the other hand, using a simulation software package with graphical animation capability can be very helpful in debugging as well as visualizing the system’s different design options.

Regardless of the modeling medium used, formulating the simulation model depends on the objectives of the study and the particular response measures at stake. Examples of study objectives include: determining the system throughput, manpower requirements, and equipment utilization. These objectives lead to the selection of the system’s response measures of interest (also known as dependent or output variables). Examples include: average hourly pallet throughput rate, average time-in-the-system for a semiconductor wafer lot, or average vehicle utilization in an AGV system.

Once the performance measures are finalized, the analyst’s task is to model the system’s inner logical workings with particular attention to the system characteristics that most directly affect the designated response measures. The system characteristics, also called, independent variables, are critical because experimenting with different values allows the analyst to optimize the system for the performance measures of interest. Examples of the independent variables include: conveyor speed, operator picking time, number of AGV vehicles, and load sortation method.

Simulation is an art form that relies on the analyst’s experience and knowledge to capture key internal interactions and independent variables that may affect critical performance measures, and then explore different options. A real system may have hundreds of hardware and logical attributes and functions, and a seasoned analyst knows how to select the right ones to capture for modeling.