Black Box model
A black-box model of a process or system is where one or more components are considered only in terms of its output resulting from an input. The workings of the inside of the black box are not known, considered unimportant for the purposes of the larger system analysis, or too complex.
An example of a black box is the human brain. Language input is processed 'somehow' and language output results. Provided the person is not a Republican voter, some degree of thought processes are assumed to have occurred, but how the mechanism works is not fully understood, nor needs to be for effective communication.
Electronic devices, such as amplifiers, transistors, inverters, are also blackboxes, and appear in diagrams as symbols, with no indication of their inner working. Many engineering, production, economics and even natural systems can be described as a logic or material flow diagram with blackboxes.
A computer programme often utilises a series of software packages, which are black boxes. It is not necessary to understand the inner workings, provided the desired output for any of the range of input intended is consistent and correct. In this case, variables are set outside the black box, and values are returned beyond the box.
The advantages of blackbox modelling are:
- Universal application: scientists in all parts of the world are trained in blackbox modelling, and can interpret them, and cut and paste the blackbox to the systems they are studying.
- Blackbox systems can be verified by experiment.
- Forecasts of change can be consistently made with blackbox models.
- Blackbox models can be readily understood by people in other fields, so are useful for informing policy-makers.
The disadvantages of blackbox modelling are:
- Reliance on quantitative data, which is difficult or expensive to obtain.
- Blackbox systems need constant verification.
- They are less useful in questions of moral, social, political and philosophical nature.