scientific article; zbMATH DE number 3854762
In writing this review, the reviewer does so not as a specialist in the theory of automata discussed in Chapter 2 of the text, but rather as someone to whom the subject matter of complex systems is of essential interest in his main area of interest, viz. decision making, and who, in reading the text, was seeking answers to the issues that complexity poses in this area. Also, in view of the complexity of even describing the mathematical framework, a general, non-mathematical review will be given. Complexity is a popular, but vague term, used to denote that, in some sense, severe difficulties can arise, from different sources, in studying, let us say, non-trivial systems. The author attempts to make the notion of complexity more precise, and to draw conclusions about qualitative properties of systems from his notions of complexity, such as the relationship between complexity and survival, the latter objective being the central one he attributes to systems. In the opinion of this reviewer this has not been really achieved, and, what has been done, is to illustrate certain features of behaviour of a selection of systems which stem from a study of the complexity of structure, time behaviour, and computation involved in a study of such systems. It is fair to say that one conclusion coming from the text, namely that the computational requirements to do anything at all are so obviously beyond our capacity to meet, that some attempt to structure our systems to make them amenable to feasible computational algorithms is absolutely necessary. Chapter 1 is a chapter dealing with the general ideas and significance of complexity, in a lay sense, and, introduces, for example, the ideas of catastrophe theory, as a means of showing that system behaviour can qualitatively change quite significantly for relatively small perturbations of certain factors. This, in itself, does not seem of necessity to be an indication of complexity in itself, but it does indicate important aspects of system behaviour. The main message of this chapter seems to be that a fruitful analysis of systems requires computable algebraic approaches, and, not for example, topological approaches, and that, moreover, the discrete time models are more natural, and more relevant to computation, than continuous time ones. Chapter 2 contains the essential mathematics. It deals with finite machines (automata) in terms of input messages, internal states, transformations, and outputs. This machine format is used in subsequent chapters to model a selection of systems, to which we will return later. The central approach is to descrribe the finite automata in the context of semi groups (non-reversible groups) of transformations, so that a machine is seen as a set of transformations linked in series or parallel fashion, where inputs from several component machines to other component machines, as well as internal states, undergo transformations. The group complexity of the machine is the smallest integer n for which the machine can be decomposed into a mixture of n finite groups and \(n+1\) finite combinatorial semi groups. Thus complexity, for the author, centres on the ability to decompose the machines in this sense, and, from a complexity point of view, is a central feature in design and control. Several results are presented concerning particular cases. This chapter is quite technical, but the remaining chapters have a general value without reading this. The remaining five chapters deal with the topics of: complexity and dynamics, structural characteristics, bounded rationality and problem solving, complexity and decision rules, and organizational decision making, where, in each case, the procedure is to look at examples in these areas as machines (automata) as outlined in Chapter 2. Computational complexity and structural complexity are the key notions throughout (these are defined), with the same general message that computable algorithms require an appropriate decomposition of the equivalent machine - although finding such decompositions is an open question, as the reviewer well knows in his own area.
- Finite automata play the repeated prisoner's dilemma
- Organisations as teams of automata
- Complexity of games and bounded rationality
- Managing complexity. Practical considerations in the development and application of ABMs to contemporary policy challenges
- scientific article; zbMATH DE number 14504 (Why is no real title available?)
- COMPLEXITY AND INFORMATION TECHNOLOGY IN DYNAMIC SYSTEMS
- scientific article; zbMATH DE number 1333472 (Why is no real title available?)
- scientific article; zbMATH DE number 835823 (Why is no real title available?)
- Coping with complexity in the management of organized systems
- Choice and complexity
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3323650)