By Shengyong Wang, University of Akron.
Simulation is unique in its ability to solve a wide variety of problems. This dexterity is derived from the equally diverse simulation methods which are available. In its essence, simulation is a numerical method, a technique involving repeated computer calculation, rather than complex analytical mathematics. By solving through this different paradigm, we are no longer held back by problem complexity, or forced to make large assumptions in solving problems which are common to most engineering problems. Specifically in industrial engineering, the most often used method is discrete event simulation. This type of simulation focuses on discrete events, of those that incorporate variables such are discrete.
As well, the events which dictate systems changes must also happen discretely as opposed to continuously. Now by discrete, we are refereeing to having values such as 2,3,4 ect. A perfect example of this type of problem would be a simple queuing system, where there is a queue of parts waiting to enter some server for processing. Discrete variables might be, the number of parts in the queue, or number parts being processed. The events which cause state changes would be like a part entering the system, finishing processing, or exiting the system. If we were to measure the queue length with respect to time, we would see the distinct discrete shape.
Another benefit of this type of simulation is the amount of high-end mature software which is available to create and run these simulations. This offers users the ability to quickly and easily model complex systems. As well, another benefit which is often overlooked is the ability to create attractive 3-d models. Often pointed out, especially by new modelers, is that for most systems, the ability to create a 3-d model is not required. This criticism is also completely valid. Still, there is another aspect of simulation which is not being considered. We are aware that simulation is used to solve problems, to probe at and understand complex systems. The end results of the simulation are entirely based on the skill of the user, input data, and accuracy of the model formulation. But there is another key aspect which must not be neglected. Belief.
As a simulator, our goal is to both answer the concerns which prompted us to create the simulation. The expansion of a model to 3-d is the best way to convince those not familiar with simulation techniques, (managers, leadership) that this model which you have created is accurate. It allows them to “see” that this model is their system, because it looks just like it. Without their buy-in, any results achieved by the model are negated, because those that the model is supposed to support with decision making are not supporting the model. We understand that the model would give the same results, whether 2-d or 3-d, and that 3-d is no more accurate. But there is no better tactic to getting the crucial buy-in that you need.