Recently, more and more articles have appeared on various websites, articles and news reports, detailing advances and breakthroughs using computer simulations. One recent feature of this was a new generation of chaos simulations. These, in essence were simulation models that worked out every possible scenario and outcome, based on the parameters the user had input, and then presented the most likely results. This is slowly becoming a popular system within computer simulations, being used by a variety of government departments and commercial companies. The question many people are asking, however, is ‘What exactly is a computer simulation?’ and ‘how can a computer program help scientific work?’
What computer simulations are ultimately all about is control. To maintain a degree of control in an otherwise uncontrollable world that we live in (and throw in some chaos for good measurement…) is something to strive for. Computer simulations help give insight into the complexity of a situation so as to help us better understand and ultimately control it as much as we possibly can. So how does that happen?
In short, a computer simulation is a computer program that is used to model and almost digitally recreate a real-life or hypothetical situation. From this, the situation can be studied and the model enhanced. For example, a simulation could model a car crash, to test safety equipment and mechanisms in new vehicles. Similarly, a simulation could be used to predict pollutants in a given area, over a set amount of time.
By running these simulation models, researchers and observers are able to retrieve key information that would otherwise either not be available, or take a lot of time, manpower, money, or a combination of all three.
Similar to analysing atmospheric and geographic data, simulations can also be used to analyse a wide variety of other parameters. One example is a simulation used to model the theories of human cognition and performance. The Adaptive Control of Thought–Rational model (ACT-R) specifies how the human brain is organised which would enable individual processing modules to produce cognition.
Modern scientific research and testing is slowly incorporating simulation science and, as a result, scientists are able to expand their research and testing further than before and complete tests faster with minimal effort. In many fields of science such as medicine, engineering, economics and sociology simulation science has become invaluable towards better understanding the complex world we live in. It can be argued that any situation or context that we can think of can be simulated and experimented with in order to gain insight into the different dynamics at play.
In a more casual context, the term ‘Computer Simulation’ can be used to describe any form of computer-generated representation, from games, to future-scenario predictions. This term, however is much broader than computer modelling, and should not be confused. Whereas computer modelling implies that all aspects of the scenario are modelled by the computer, computer simulations can also generate inputs from simulated users.
An example of this would be a flight simulator, which can run machines, as well as flight software. The system can run ‘artificial intelligence’ entities, in this example, other aircraft in the simulations, Air Traffic Controllers, various airport vehicles etc. Yet also have the capability to support a completely manual interaction, as in the player controlling an aircraft, amongst the mass of AI. These can then interact and observe the inputs and actions of the user. For example, if the user chose to deviate from the set path, the AI entities would respond to the action in a logical, lifelike manner. For example, Air Traffic Controllers would request an explanation of the deviation; other aircraft would move to avoid the user and so on. Whereas a model requires a set of parameters at the start, and from then on, the scenario is modelled by the computer, a simulation can support and adapt to constant changes in the variables.
A recent example of computer simulations proving beneficial for the wider community featured during the H1N1 flu outbreak of 2009/10. The State of Ohio used computer simulations, developed in close consultation with the federal Centres for Disease Control, to predict what areas would have increased chances out outbreaks. Using this data, medical supplies and vaccines were distributed accordingly. It is estimated that this system prevented more than 310,000 cases of flu that would have resulted in 64 deaths and 1,400 hospitalisations. According to the study, the hospitalisations alone would have cost about $8.4 million.
