| |
| |
| |
| Overview |
|
One question we are asked on occasion is what is the difference between open and closed systems. It seems that some individuals place great importance on the distinction between these two system types, and exert considerable effort in order to design processes within a closed system context. At face value, we at the Artige Company consider all systems to be open, so we do not spend much thought with closed systems. The analysis below will explain what open and closed systems are, and what the difference is between them. We will also explain why the Artige Company focuses on open systems. |
| |
| |
| |
| Analysis |
|
Since this monograph discusses open and closed systems, we will need to define what a system is in the first place. At the Artige Company, we like to use the definition where a system is composed of a transformation. That is, an activity that takes inputs and creates outputs from them. This transformation mechanism has the ability to monitor its actions, and possibly adjust the transformation mechanism, through a process called feedback. This definition is demonstrated in Figure 5.1 below, which originally appeared in an XML textbook written by one of the Artige consultants. |
| |
| |
|
 |
| |
| |
|
There is another distinction that we like to make when we talk about a system, one that can be confusing at times. We talk about both systems and processes, and we consider these two artifacts to be different from each other. Actually, there is a parent - child relationship between them. A process is compromised of one or more systems. A process is the token that human beings like to use as shorthand for "getting things done". The idea is to have a proxy that can stand in to summarize all of the transformations and their interactions that will produce a desired result from an assemblage of known inputs without getting into the explicit detail of what each and every system entails in order for the process to be considered successful. |
| |
| |
|
Frequently you will find that we at the Artige Company focus on the transformation when considering a system, as that is where the action takes place. The inputs and outputs are the inanimate articles, and do not themselves provide any action. |
| |
| |
|
With the definitions out of the way, we can discuss the concept of open and closed systems. This is where one brings the ideas of boundaries and delineation into the discussion. A closed system is one where all of the inputs and outputs have been identified, and are not affected by any other factors. For this concept to be realized, there must be other inputs and outputs that are present, and those persons responsible for the closed system in question are drawing a line that delineates the inputs and outputs between their closed systems and the rest of the universe. |
| |
| |
|
An open systems is one, where conversely to the closed system, all of the inputs and outputs have not been identified, plus it is possible [most likely a result of the lack of complete input and output identification] that inputs and outputs are interacting with the open system in question, possibly with other systems which may or may not have been identified as being part of a process. |
| |
| |
| |
|
Delineation vs. Categorization |
| |
| |
|
It may be apparent to the reader that a closed system is an academic device, useful only for explaining system theory and how systems function. In actual practice one will not find any system that is closed. It would be difficult to prove that one has located and isolated all of the inputs and outputs of a system. Thanks to the reductionist tools that we acquired through the scientific revolution, any input and output can be decomposed into smaller components. Then it will be necessary to show that the newfound inputs and outputs are not being used in another system. This exercise becomes open-ended, which makes it difficult to prove that a system can be entirely and absolutely delineated, so that it could be deemed "closed". This is just not going to happen using a reductionist methodology. |
| |
| |
|
It is our contention that since all systems are open, it then becomes imperative that a method of modeling systems is devised that recognizes the fact that the system designer will be focusing on a specific subset of inputs and outputs, and that other inputs and outputs exist that might affect the outcome of the system being designed. For us, such a method is categorization. This concept is not new, and for most instances, systems practitioners have already been using this methodology to some degree. For instance in the field of decision sciences, categorization has been used to provide guidance in making business decisions, by converting apples-to-oranges problem domain into an oranges-to-oranges (or apples-to-apples) domain. |
| |
| |
|
Categorization is the task of surveying the system in question, in an attempt to identify all of the inputs and outputs for the transformation. These inputs and outputs are then classified by the impact they have on the system. In most instances, one will focus on the inputs and outputs that have the most impact on the system. The popular "80/20 rule" will frequently be the informal criteria when determining when it is no longer constructive to continue seeking out inputs and outputs. Note that a categorization exercise does not have a fixed formula, no one-and-only method of determining inputs and outputs. There are always exceptions to the common experiences of systems designers, where an insignificant input or output can still have a discernable impact on the success of a system. The "butterfly effect" would be the typical example to drawn on for such an instance. |
| |
| |
|
Another item of concern for the system designer is the interaction of systems on each other. This is not a consideration in a closed system, which becomes is a source of wonderment for those closed system designers that experience unexpected outcomes. One of the most common explanations for unexpected input and outputs in system design has to do with systems running in parallel at different levels that are using each others inputs and outputs. For instance, a production machine will be scheduled by a production management system, while its operators are scheduled through a manpower system. This need to satisfy criteria simultaneously is explained in further detail in our BPDA article. That article explains the concept of mapping inputs and outputs of systems running at various levels. Suffice it to say in this article that system design cannot ignore the systems in its environment without dire effects. |
| |
| |
| |
| Conclusion |
|
This article explained what compromises a system, and explained what closed and open systems are. It was noted that a closed system is a tool useful only as an academic exercise, in order to explain functionality of a system. There are no true closed systems. As far as the Artige Company is concerned, all systems are open. Problem one has to deal with then, can you locate all of the inputs and outputs? So be aware that a system can be modeled that focuses only on the inputs and outputs that have a major impact on the outcome of the system. It is still incumbent upon the system designer to consider the implications of not taking all inputs and outputs into account. The lack of our current ability to identify inputs and outputs is one thing. To make business decisions by purposefully ignoring the interaction of systems with each other is just asking for defects and a lack of success to plague a business on a daily basis. |
Legal