How can we think the complex--论文代写范文精选

这些相互作用使系统的行为很难跟踪，如果是非线性的相互作用,系统的描述不能通过元素的描述。我们可以提到一个细胞,乃至一个社会，他们都由许多元素组成。然而,跟踪系统的功能和属性不是一项容易的任务。下面的essay代写范文进行讲述。

Summary
In this chapter we want to provide philosophical tools for understanding and reasoning about complex systems. Classical thinking, which is taught at most schools and universities, has several problems for coping with complexity. We review classical thinking and its drawbacks when dealing with complexity, for then presenting ways of thinking which allow the better understanding of complex systems. Examples illustrate the ideas presented. This chapter does not deal with specific tools and techniques for managing complex systems, but we try to bring forth ideas that facilitate the thinking and speaking about complex systems.

Itroduction
Complexity Giving a sharp definition of complex system is hard, since the term is used in such a wide variety of contexts. Because of this, we will only give a notion, to have a better idea of what we are speaking about. A complex system is usually hard to describe (although simple systems can also be, therefore not to confuse complex with complicated), because it consists of several elements that interact with each other (see Bar-Yam (1997) for a detailed introduction). These interactions make the global behaviour of the system hard to track, in terms if its elements. If the interactions are nonlinear, then the description of the system cannot be reduced to the description of the elements. The properties present at the system level not present at the element level are called emergent. Examples of complex systems are everywhere. We can mention a cell, a society, an economy, an ecosystem, the Internet, the weather, a brain, a city. 

They all consist of many elements, and the functions and properties ofthe systemare a result of the elements’ interactions. Nevertheless, tracking functions and properties of the systems to single elements or interactions is not an easy task. There are several measures of complexity, useful in different contexts: information, social, economic, biological, etc. Also, we cannot draw a sharp boundary between simple and complex systems. What we can do is to compare according to an agreed frame of reference and say: this system is more complex than that one. Overall, we can say that the complexity of a system scales by its number of elements, by the number of interactions among them, by the complexities of the elements and by the complexities of the interactions. This is a recursive measure, but it can be general enough to be applied in different contexts. We just have to note that the more interactions, the more complex a system will be. And the more elements, the higher the complexity of the system. For example a firm will be more complex the more employees it has. However, for the same number of employees, a firm will be more complex than other if people need to interact more between them, since more coordination will be required. As we can see, complex systems tend to be harder to manage than simple ones. The main reason for this is that we know how to think about simple systems, when we are only beginning to understand the complex.

Causality 
Classical causality is very useful when A affects B, but B does not affect A too much. For example, the gravity of the sun affectsthe earth quite a lot, causing itsrotation around it, whereas the effect of earth’s gravity in the sun is neglectible. Then we can conclude that the sun causes the earth to orbit around it. We can say that classical causation is a relation where we ignore the effect of the “consequent” on the “cause”. Classical logic is based on these premises, and it has served human kind tremendously: it enables us to make accurate predictions (Heylighen, 1989). However, it seems that we run into problems when we cannot explain the behaviour of a system when its elements tend to affect each other simultaneously. It is a kind of chicken-andegg problem. We have A affecting B, but simultaneously, B affecting A. Moreover, if there is some nontrivial time delay between these effects (Gershenson, 2002b; Gershenson et al., 2003), it becomes very hard to make accurate predictions, because we do not know who will affect who first, and this can have drastic effects on the dynamics of the system. The more interactions and/or elements we have in a system, the higher the probability will be that two elements will affect each other. Therefore, classical causality runs into trouble when trying to understand or predict a complex system. We need to use a different way of thinking to understand them.

Nonlinearity 
A common characteristic of models of complex systems, is that they are nonlinear. This means that the elements of a system interact in ways that are more complex than additions and subtractions. In a linear system, we just add the properties of the elements, and we can deduce and predict the behaviour of the system. Nevertheless, when there are many interactions, and these are nonlinear, small differences multiply overtime, yielding often chaos and unpredictability. In a nonlinear system, causes are not directly proportionalto their effects. Big changes can have little or no effect, while small changes can have drastic consequences. This makes complex systems to be not completely predictable.

Open and Closed Systems 
One implication of classical causality is the tendency to model systems as closed. This means that we neglect the effects of elements outside our model. This is understandable, because we model only what we see, which is finite. Again, this brings problems when the system complexity is such that we cannot ignore unpredicted causes outside our model. With some experience we can see that all systems (abs)are open (e.g. affected by external gravitational or electromagnetic forces), although in practice we can model simple systems as closed. This is because we can neglect small perturbations, since they do not affect the behaviour of the simple system. 
But we cannot do this with complex systems. Unpredictable events very probably will come into existence when we have complexity. Moreover, small perturbations can propagate to produce drastic changes in the system. This is natural. A possible solution: imitate nature. Nature adapts to unpredicted changes and events. Many tools mentioned above take inspiration from nature to make artificial complex systems adaptive. Another example is ant colony optimization: its algorithms are inspired in the self-organization of the social insects. Having this in mind, we should try to model all complex systems as open systems, in the sense that there will be unknown factors bringing things in and out of our system. The standard way of achieving this is introducing noise. In other words, we include in our model small random fluctuations. Noise allows us to observe and measure how robust a system is. Actually, many systems require noise to be robust and stable. An example can be software development. Within a classic framework, there is a requirement, and the system should comply with it. However, if we are dealing with a complex software system, the requirements can change unexpectedly with a higher probability, for whatever reasons. If the software were developed fromthe beginning as an open system, changes and extensions are already expected, so the systemcan adapt much better. Object-oriented system engineering has been a big step towards this, since it allows the easy reutilization of code. Therefore, in an idealscenario,smallmodules can be adjusted without the need of reimplementing the rest of the system. Further research is being carried out for having modules that can adjust themselves to unexpected changes. Another example can be a company. It can follow a strict business model. However, if the company is in an unpredictable environment, as all actual companies are, that business model should better expect the unpredictable, to be ready to adapt to unforseen changes, such as new competition, new market opportunities, or new products in the market. It is by realizing systems as open that we can be ready and face unforseen changes. We cannot do this with classical thinking, because this one assumes that the world is predictable. Well, with complex systems we have seen that it is not, and we have to change our ways of thinking to cope with it better. 8. 

Conclusions 
We still do not understand complexity very well. There is much to be done and explored in this direction. Our culture now is immersed and surrounded by complexity, and we have no other option than to face it. But facing complexity forces us to change our ways of thinking. We have presented some philosophical tools that allow us to better understand and speak about complexity. One of the main things we have to be conscious about when we are dealing with complex systems is that they are not completely predictable, even if we know how they function. The fact that we understand a problem does not mean that we will be able to solve it. Yet what we can, and should do, is to be prepared to deal with the unexpected events that complexity most certainly will bring forth. We should not try to determine the behaviour of a complex system, but to expect certain possibilities. And we should try to be able to adapt when the unexpected comes. Because then we are ready to expect the unexpected.

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。（essay代写)
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X（essay代写)