Systemic analysis is a process of receiving answer to the question “Why is the overall goal of the system fulfilled (not fulfilled)?” The notion of “systemic analysis” includes other two notions: “system” and “analysis”. The notion of “system” is inseparably linked with the notion of the “goal/purpose of the system”. The notion “analysis” means examination by parts and arranging systematically (classification). Hence, the “systemic analysis” is the analysis of the goal/purpose of the system by its sub-goals (classification or hierarchy of the goals/purposes) and the analysis of the system by its subsystems (classification or hierarchy of systems) with the view of clarifying which subsystems and why can (can not) fulfill the goals (sub-goals) set forth before them. Any systems perform based on the principle “it is necessary and sufficient” which is an optimum control principle. The notion “it is necessary” determines the quality of the purpose, while the notion “is suficient” determines its quantity. If qualitative and quantitative parameters of the purpose of the given system can be satisfied, then the latter is sufficient. If the system cannot satisfy some of these parameters of the goal, it is insufficient. Why the given system cannot fulfill the given purpose? This question is answered by systemic analysis. Systemic analysis can show that such-and-such object “consists of... for…”, i.e. for what purpose the given object is made, of what elements it consists of and what role is played by each element for the achievement of this goal/purpose. The organic-morphological analysis, unlike systemic analysis, can show that such-and-such object “consists of... “, i.e. can only show of which elements the given object consists. Systemic analysis is not made arbitrarily, but is based on certain rules. The key conditions of systemic analysis are the account of complexity and hierarchy of goals/purposes and systems.
Complexity of systems. It is necessary to specify the notion of complexity of system. We have seen from the above that complexification of systems occurred basically for the account of complexification of control block. At that, complexity of executive elements could have been the most primitive despite the fact that control block at that could have been very complex. The system could contain only one type SFU and even only one SFU, i.e. to be monofunctional. But at the same time it could carry out its functions very precisely, with the account of external situation and even with the account of possibility of occurrence of new situations, if it had sufficiently complex control block. When the analysis of the complexity of system is made from the standpoint of cybernetics, the communication, informo-dynamics, etc. theories the subject discussed is the complexity of control block, rather than the complexity of the system. Note should be taken of that regardless of the degree of the system complexity two flows of activity are performed therein: information flow and a flow of target-oriented actions of the system. Information flow passes through the control block, whereas the flow of target-oriented actions passes through executive elements. Nevertheless, the notion of complexity may also concern the flows of target-oriented actions of systems. There exist mono- and multifunctional systems. There are no multi-purpose systems, but only mono-purpose systems, although the concept of “multi-purpose system” is being used. For example, they say that this fighter-bomber is multi-purpose because it can bomb and shoot down other aircrafts. But this aircraft still has only one general purpose: to destroy the enemy's objects. This fighter-bomber just has more possibilities than a simple fighter or simple bomber. Hence, the notion of complexity concerns only the number and quality of actions of the system, which are determined by a number of levels of its hierarchy (see below), but not the number of its elements. Dinosaurs were much larger than mammals (had larger number of elements), but have been arranged much simpler. The simplest system is SFU (Systemic Functional Unit). It fulfills its functions very crudely/inaccurately as the law that works is the “all-or-none” one and the system's actions are the most primitive. Any SFU is the simplest/elementary defective system and its inferiority is shown in that such system can provide only certain quality of result of action, but cannot provide its optimum quantity. Various SFU may differ by the results of their actions (polytypic SFU), but they may not differ either (homotypic SFU). However, all of them work under the “all-or-none” law. In other words, the result of its action has no gradation or is zero (non-active phase), or maximum (active phase). SFU either reacts to external influence at maximum (result of action is maximum - “all”), or waits for external influence (the result of action is zero - “none”) and there is no gradation of the result of action. Each result of SFU action is a quantum (indivisible portion) of action. Monofunctional systems possess only one kind of result of action which is determined by their SFU type. They may contain any quantity of SFU, from one to maximum, but in any case these should be homotypic SFU. Their difference from the elementary system is only in the quantity of the result of action (quantitative difference). The monofunctional system may anyway perform its functions more accurately as its actions have steps of gradation of functions. The accuracy of performance of function depends on the value of action of single SFU, the NF intensity and the type of its control block, while the capacity depends on the number of SFU. The “smaller” the SFU, the higher the degree of possible accuracy is. The larger the number of SFU, the higher the capacity is. So, if the structure of the system's executive elements (SFU structure) is homotypic, it is then multifunctional and simple system. But at that, its control block, for example, may be complex. In this case the system is simple with complex control block. The multifunctional system is a system which contains more than one type of monofunctional systems. It possesses many kinds of result of action and may perform several various functions (many functions). Any complex system may be broken down into several simple systems which we have already discussed above. The difference of multifunctional system from the monofunctional one is that the latter consists of itself and includes homotypic SFU, while complex system consists of several monofunctional systems with different SFU types. And at that, these several simple systems are controlled by one common control block of any degree of complexity. The difference between monofunctional and multifunctional systems is in the quantity and quality of SFU. In order to avoid confusion of the complexity of systems with the complexity of their control block, it is easier to assume that there are monofunctional (simple) and multifunctional (complex) systems. In this case the concept of complexity of system would only apply to control block. In monofunctional system control block operates a set of own SFU regardless of the degree of its complexity. In multifunctional system control block of any degree of complexity operates several monofunctional subsystems, each of which has its SFU with their control blocks. It is complexity of control block that stipulates the complexity of the system, and not only the type of system, but the appurtenance of the given object to the category of systems. The presence of an appropriate control block conditions the presence of a system, whereas the absence of (any) control block conditions the absence of a system. Systems may have control blocks of a level not lower than simple. The full-fledged system can not have the simplest/elementary control block, whereas the SFU can.
So, the system is an object of certain degree of complexity which may tailor its functions to the load (to external influence). If its structure contains more than one SFU, the result of its action has the number of gradations equal to the number of its SFU or (identically) the number of quanta of action. The number of the system's functions is determined by the number of polytypic monofunctional systems comprising the given system. In former times development of life was progressing towards the enlargement of animal body which provided some kind of guarantee in biological competition (quantitative competition during the epoch of dinosaurs). But the benefits has proven doubtful, the advantages turned out to be less than disadvantages, that is why monsters have died out. This is horizontal development of systems. If they differ in quality it is tantamount to the emergence of new multifunctional systems. Such construction of new systems is the development of systems along the vertical axis. The example of it is complexification of living organisms in process of evolution, from elementary unicellular to metazoan and the human being. What can be done by man can not be done by a reptile. However, what can be done by reptile can not be done by an infusorian (insect, jellyfish, amoeba, etc.). Complexification of living organisms occurred only for one cardinal purpose: to survive in whatever conditions (competition of species). Since conditions of existence are multifarious, the living organism as a system should be multifunctional. The character of a new system is determined by the structure of executive elements and control block features. If there is a need to extend the amplitude or the capacity of system's performance the structure of executive elements should be uniform. To increase the amplitude of the system's performance all SFU are aligned in a sequential series, while to increase the capacity - in a parallel series depending on the required quantity of the result of action (amplitude or capacity at the given concrete moment). Polytypic SFU have different purposes and consequently they have different functions. The differences of SFU stipulate their specialization, whereby each of them has special function inherent in it only. If the structure of any system comprises polytypic SFU, such system would be differentiated, having elements with different specialization. In systems with uniform SFU all elements have identical specialization. Therefore, there is no differentiation in such system. So, the concept of specialization characterizes a separate element, whereas the concept of differentiation characterizes the group of elements. The number of SFU in real systems is always finite and therefore the possibilities of real systems are finite and limited, too. Resources of any system depend on the number of SFU comprising its structure in the capacity of executive elements. The pistol may produce as many shots as is the number of cartridges available in it, and no more than that. The less the number of SFU is available in the system, the smaller the range of changes of external influence can lead to the exhaustion of its resources and the worse is its resistance to the external influence. By integrating various SFU in more and more complex systems it is possible to construct the systems with any preset properties (quality of the result of action) and capacities (amount of quanta of the result of action). At that, the elements of systems are the systems themselves, of a lower order though (subsystems) for these systems. And the given system itself may also be an element for the system of higher order. This is where the essence of hierarchy of systems lies.
Hierarchy of goals/purposes and systems. The more complex the system, the wider the variety of external influences to which it reacts. But the system should always produce only specific (unique, univocal) reaction to certain influence (or certain combination of external influences) or specific series of reactions (unique/univocal series of reactions). In other words, the system always reacts only to one certain external influence and always produces only one specific reaction. But we always see “multi”-reactive systems. For example, we react to light, sound, etc. At the same time we can stand, run, lay, eat, shout, etc., i.e. we react to many external influences and we do many various actions. There is no contradiction here, as both the purposes and reactions may be simple and complex. The final overall objective of the system represents the logic sum of sub-goals/sub-purposes of its subsystems. The goal/purpose is built of sub-goals/sub-purposes. For example, the living organism has only one, but very complex purpose - to survive, by all means, and for this purpose it should feed. And for this purpose it is necessary to deliver nutriment for histic cells from the external medium. And for this purpose it is necessary first to get it. And for this purpose it is necessary to be able to run quickly (to fly, bite, grab, snap, etc.). Thereafter it is necessary to crush it, otherwise it won't be possible to swallow it (chewing). Then it is necessary to “crush” long albumen molecules (gastric digestion). Then it is necessary to “crush” the scraps of the albumen molecules even to the smaller particles (digestion in duodenum). Then it is necessary to bring in the digested food to blood affluent to intestine (parietal digestion). Then it is necessary... And such “is necessary” may be quite many. But each of these “is necessary” is determined by a sub-goal at each level of hierarchy of purposes. And for every such sub-goal there exists certain subsystem at the respective level of hierarchy of subsystems. At that, each of them performs its own function. And in that way a lot of functions are accumulated in a system. However, all this hierarchy of functions is necessary for one unique cardinal purpose: to survive in this world. Any object represents a system and consists of elements, while each element is intended for the fulfillment of respective sub-goals (subtasks). The system has an overall specific goal and any of its elements represents a system in itself (subsystem of the given system), which has its own goal (sub-goal) and own result of action. When we say “overall specific goal” we mean not the goals/purposes of elements of the system, but the general/overall/ purpose which is reached by means of their interactions. The system has a goal/purpose which is not present in each of its element separately. But the overall goal of the system is split into sub-goals and these sub-goals are the purposes of its elements anyway. There are no systems in the form of indivisible object and any system consists of the group of elements. And each element, in turn, is a system (subsystem) in itself with its own purpose, being a sub-goal of the overall goal/general purpose/. To achieve the goal the system performs series of various actions and each of them is the result of action of its elements. The logic sum of all results of actions of the system's subsystems is final function - the result of action of the given system. Thus, one cardinal purpose determines the system, while the sub-goal determines the subsystem. And so on and so forth deep into a hierarchy scale. The goal/purpose is split into sub-goals/sub-purposes and the hierarchy of purposes (logically connected chain of due actions) is built. To perform this purpose the system is built which consists of subsystems, each of which has to fulfill their respective sub-goals and capable to yield necessary respective result of action. That is how the hierarchy of subsystems is structured. The number of subsystems in the system is equal to he number of subtasks (subgoals) into which the overall goal is broken down. For example, the system is sited at a zero level of hierarchy, and all its subsystems are sited at a minus one, minus two, etc. levels, accordingly. The order of numeration of coordinates is relative. It means that the given system may enter the other, larger system, in the capacity of its subsystem. Then the larger system will be equalized to zero level, whereas the given system will be its subsystem and sited at a minus one level. The hierarchy scale of systems is built on the basis of hierarchy of goals/purposes. Target-specific actions of systems are performed by its executive elements, but to manage their target-oriented interaction the interaction of control block of the system with control blocks of its subsystems is needed. Therefore, the hierarchy scale of systems is, as a matter of fact, a hierarchic scale of control blocks of systems. This scale is designed based on a pyramid principle: one boss on top (the control block of the entire system), a number of its concrete subordinates below (control blocks of the system's subsystems), their concrete subordinates under each of them (control blocks of the lower level subsystems), etc. At each level of hierarchy there exist own control blocks regulating the functions of respective subsystems. Hierarchical relations between control blocks of various levels are built on the basis of subordination of lower ranking blocks to those of higher level. In other words, the high level control block gives the order to the control blocks of lower level. Only 4 levels of hierarchy, from 0 to 3rd, are presented. The count is relative, whereby the level of the given system is assumed to be zero. The counting out may be continued both in the direction of higher and lower (negative) figures/values. The notions of “order” and “level” are identical. The notions of “system” and “subsystem” are identical, too. For example, instead of expression “a subsystem of minus second-order” one may say “a system of minus second-level”. And although a zero level is assumed the level of the system itself, the latter may be a part of other higher order system in the capacity of its subsystem. Then the number of its level can already become negative (relative numeration of level). Elements of each hierarchic level of systems are the parts of system, its subsystems, the systems of lower order. Therefore, the notions “part”, “executive element”, “subsystem”, “system” and in some cases even “element” are identical and relative. The choice of term is dictated only by convenience of accentuating the place of the given element in the hierarchy of system. The notion of hierarchic scale (or pyramid principle) is a very powerful tool and it embodies principal advantage of systemic analysis. Systemic analysis is impossible without this concept. Both our entire surrounding world and any living organism consist of infinite number of various elements which are relating to each other in varying ways. It is impossible to analyze all enormous volume of information characterizing infinite number of various elements. The concept of hierarchy of systems sharply restricts the number of elements subjected to the analysis. In the absence of it we should take into account all levels of the world around us, starting from elementary particles and up to global systems, such as an organism, a biosphere, a planet and so on. For global evaluation of any system it is sufficient to analyze three levels only: the global level of the system itself (its place in the hierarchy of higher systems); the level of its executive elements (their place in the hierarchy of the system itself); the level of its control elements (elements of control block of the system itself). To evaluate the system's function it is necessary to determine the conformity of the result of action of the given system with its purpose - due result of action (global level of function of the system), the number of its subsystems and the conformity of their results of action with their purposes - due results of their action (local functional levels of executive elements) and evaluate the function of elements of control. In the long run the maximum level of function of system is determined by the logic sum of results of actions of all subsystems comprising its structure and optimality of control block performance. Abiding by the following chain of reasoning: “the presence of the goal/purpose for implementation of any specific condition, the presence of qualitative or quantitative novelty of the result of action, the presence of a control (block) loop” it is possible to single out elements of any concrete system, show its hierarchy and divide cross systems in which the same elements perform various functions. Systems work under the logical law which main principle is the fulfillment of condition “... if..., then….”. In this condition “if ..” is the argument (purpose), while “then...” is the function (the result of action). This condition stipulates determinism in nature and hierarchy scale. Any law, natural or social, requires implementation of some condition and the basis of any condition is this logical connective “... if..., then…” At that, this logical connective concerns only two contiguous subsystems on a hierarchic scale. The argument “... if” is always specified by the system which is on a higher step, whereas the function “then…” is always performed by the system (subsystem) sited immediately underneath, at a lower step of a hierarchic scale. Actions of elements per se and interaction between the elements may be based on the laws of physics or chemistry (laws of electrodynamics, thermodynamics, mathematics, social or quantum laws, etc.). But the operation of control block is based only on the logical laws. And as far as control block determines the character of function of systems, it is arguable that systems work under the logic laws. Sometimes in human communities the “bosses” would imagine they may govern/control/ at any levels, but such type of management is the most inefficient one. The best type of management is when the director (the control block of multifunctional system) controls/manages/ only the chiefs of departments (control blocks of monofunctional systems), sets forth feasible tasks before them and demands the implementation thereof. At that, the number of its “assistant chiefs” should not exceed 7±2 (Muller's number). If some department does not implement its objectives, it means that either the departmental management (control block of a subsystem) is no good because has (a) failed to thoroughly devise and distribute the tasks between the subordinates (the SFU), or (b) has inadequately selected average executives (SFU), or (c) impracticable goal has been set forth before the department (before system), or (d) the director himself (control block of the system) is no class for the management. In such cases the system's reorganization is necessary. But if the system is well elaborated and performs normally there is no sense for the director to “pry” into the department's routine affairs. A chief of department is available for this purpose. The decision of the system reorganization is only taken when the system for some reason cannot fulfill the objective (system crisis). In the absence of crisis there is no sense in reorganization. For the purpose of reorganization the system changes the structure of its executive and control elements both at the expense of actuation (de-actuation) of additional subsystems and alteration of exit-entry combinations of these elements. In such cases skipping of some steps of hierarchy may occur and the principle “vassal of my vassal is not my vassal” violated. This is where the essential point of the system reorganization lies. At the same time, part of elements can be thrown out from the system as superfluous (that's how at one time we lost, for example, cauda and branchiae), while other part may be included in the system's structure or shifted on the hierarchy scale. But all that may only happen in process of the system reorganization proper. When the process of reorganization comes to an end and the reorganized system is able of performing the goal set forth before it (i.e. starts to function normally), the control law of “vassal of my vassal is not my vassal” is restored.
Ñòðàíèöû: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
|