Structuredness (in Combination with the Internal Heterogeneity of the System, its openness, Functionality, Emergence, and Purposefulness)

The model of the system structure is defined as a list of significant links between parts of the system. The keyword here is the evaluative term “essential”. This concept includes the following:

• 1. information about this relationship is necessary for successfully achieving the goal with the help of the created model;
• 2. the information content of this relationship far exceeds the importance of other relationships recognized as “insignificant” and therefore not included in the model;
• 3. the aggregate information about all significant relationships is sufficient for the successful achievement of the goal.

In fact, the model of the structure arises as a result of purposeful refinement of the composition model: for each selected part of the system, a black box model is built, the inputs and outputs of which are significant connections of this part with others.

It is appropriate to start by considering the simplest (at first glance) element of the structure model — a separate tie between two fixed elements of the composition model. An important feature of the relationship is often its direction, reflecting the fact that the interaction of two entities includes their effects on each other. There are situations when the action of one on the other is equal to the action of the second on the first (as in mechanics — “action is equal to counteraction”); however, often one of the effects is superior in strength to the other, and then in the working model of the situation remains only the most powerful (as in the model of relations between the boss and the subordinate, or in the model of relations between cause and effect). Consequently, the system model often looks like a graph with unidirectional (causal) connections between pairs of vertices.

In many cases, the nature of the dynamic processes in the system depends on the kind of resource, with what intensity comes through this communication channel, and what are the possibilities of regulating this intensity. Then, there is a need to provide information about the composition and structure of the connection itself, and to make appropriate changes and additions to the structural scheme of the entire system. Within this approach, the functioning of the system is represented as the flow of resources (substances, energy, information) through channels connecting parts of the system, each of which carries out certain (quantitative and/or qualitative) transformations of the resources received as inputs, and sends the results to other parts through output channels. At the same time, there are three typical components of the model of the composition of the connection (the first is inherent in any connection, the other two are present only in some connections):

• • the communication channel (“flow”, “pipe”, “pathway”, etc.), the maximum possible flow rate of the resource through the channel is called its “capacity”;
• • flow regulator (“valve”, “tap”), which allows you to change the flow rate in the channel between zero and the capacity;
• • a reservoir (“storage pool” of a resource), in which the volume of the resource is characterized by a “stock” (or “level”) and depends on the ratio between its capacity and the speed of its input and output flows.

As a result, the system composition model includes not only its own functional parts but also the specified parts of each link (see Figure 7.5). Taking them into account is very important while considering the dynamics of processes in the system. (This will be discussed in the next paragraph.)

Further, the system can be considered a sequential chain of connected parts. The action of one entity on another can be carried out not only directly but also through the impact on third entities that are associated (directly or indirectly) with both. The chains of such intermediate entities can be arbitrarily long, and the result of the impact through a particular chain can be both positive and negative. If there are several such chains between the two entities (remember the example of the effect of smoking on the heart), the final effect is determined by the predominance of one over the other.

Another, and very important, feature of the sequences of directional ties is that the chain of pairwise influences, which began from this object, can ultimately be closed to itself. This phenomenon is called feedback. In this case, the return effect on the original source can contribute to the initial process and strengthen it (in this case, the feedback is called positive or stimulating), but can also weaken or oppress it (and then the feedback is called negative or stabilizing). The presence of feedbacks leads to a wide variety of system behaviors, which is the subject of the theory of system dynamics. (We will discuss this issue later when considering the dynamic properties of systems.)

By considering individual links and their linear sequences, we proceed to the description of the features of the whole set of links — the model of the structure of the system. Together, the composition and structure models form a structural scheme of the system, usually represented by a graph consisting of “nodes” or “vertices”

FIGURE 7.5 Elements of the resource flow.

(image of parts) and “edges” or “arcs” (image of links between parts). The direction of relationship is indicated by arrows on the edges (oriented graphs); the difference in the quality of the relationship is sometimes displayed by multicolored arcs (painted, or colored graphs) or other distinguishing features; for example, J. Forrester [3] displayed the flows of different types of resources in economic systems by arrows of different configurations (see Figure 7.6).

Duality of graphs, in which either parts or their functions can be considered as vertices, allows describing both the static and dynamic structural scheme of the system. This model enables the representation of a wide variety of systems. Their block diagrams form a wide range of different graphs, from a linear chain to a complete graph, in which each vertex is connected to all the others (Figure 7.7).

On graph models, it is convenient to consider the different features of systems related to their structure. Let’s consider some variants of system structure models.

One of the aspects that provide specifics to the structural scheme of the system is the reliability of the system, that is, its proper functioning, despite the failure

FIGURE 7.6 Flows of various resources presented by different arrows.

FIGURE 7.7 Examples of full graphs.

of some element(s). Increasing the reliability of the system is possible by introducing redundancy in the system structure. In technical systems, this is often done by duplication or parallel execution of several elements of the same function; in information transmission systems in the presence of noise, redundancy is introduced into the transmitted signal when it is encoded (the simplest example of error-correcting coding — multiple repetition of the same message); in organizational systems, the necessary redundancy is usually introduced in the form of reserve (e.g., the posts of the president and vice-president, chief executive officer and his deputy).

Hierarchy has a special place among structural schemes. This term refers to a multilevel tree-like structure diagram in which each element is associated with only one top-level element and with several lower-level elements (see Figure 5.26, Part II). This structure gives the system the properties with the results that the vast majority of natural and artificial (man-made) systems are organized hierarchically. The specific features of ideal hierarchical structure are as follows:

• • the system consists of a certain set of parts, the internal connections of the elements in each of them are stronger than the external ones, that is, the boundaries between the parts are natural, and each part performs its function, that is, has its own, imputed to its purpose;
• • the goals of the parts are agreed among themselves in such a way that their joint implementation ensures the fulfillment of the system’s goal, of which they are parts. The goals of the parts in the hierarchical structure form a tree of goals, with each element playing a dual role: on the one hand, it is the goal in relation to the subgoals of the lower level, on the other, it is a means (subgoal) of achieving the goal of the upper-level element.

Many natural and artificial systems have a hierarchical structure because this structure allows simply building up hierarchy levels to create much more complex system of relatively simple elements, while maintaining harmonization of the objectives of the parts for the purpose of the entire system. Hierarchy is a very effective means of overcoming the complexity of large dimensions.

Perhaps the main advantage of hierarchy is that it allows the storage and constant processing of huge amounts of information necessary for the existence, growth, and development of the entire system in a changing environment, with a significant limitation that each element in the structure can only work with limited amounts of information.

For example, the number of subordinates to a manager is limited by the purely psycho-physiologic limits of a person’s ability to keep in mind the activities of several subordinates at once simultaneously (empirically established in psychology rule “seven plus or minus two”). If you need to coordinate the actions of more subordinate workers, the way out is to create a hierarchy. In natural physical systems that are expanding in size, hierarchical organization is often realized in the form of spiral structures (shells of some mollusks; vortices, hurricanes, and tornadoes; in mollusks, whirlwinds, hurricanes, and tornadoes; the shape of the Milky Way and other galaxies in the universe).

A common variant of system hierarchy in nature is the fractal structure, with the same rules for combining elements at all levels of the organization (in branching plants, geological structures, in the organization of matter from elementary particles to macrocosm).

In dealing with any part of the problem, consideration of the situation may focus on the different neighborhoods of that part in the overall structure of the entire system, taking into account the different degrees of interaction of that part with other related parts. For example, if you have a disease of any organ, the doctor may prescribe treatment aimed at changing the state of only the organ, or its interaction with other subsystems of the same level of hierarchy (e.g., with the cardiovascular system); however, the doctor can also pay attention to the type of your psyche and lifestyle (related to higher levels of hierarchy), or the molecular structure of cells of the diseased organ (which is at the lower level of the structural hierarchy), and even gene alterations in the molecular structure of your DNA (in hereditary diseases).

Although hierarchical systems evolve from the lowest level, and the original purpose of the upper levels of the hierarchy is to help the lower levels achieve their goals, in real life, for various reasons, there are deviations from the rules of the ideal hierarchy, changing the characteristics of the system as a whole. Social systems are characterized by deviations because people (as goal-setting subjects) have some goals at odds with those dictated by the rules of the hierarchy. Among such situations, the most common are the following:

• • people belonging to different branches and levels of hierarchy often establish and use for their own benefit connections that are not provided by the ideal hierarchy (e.g., nepotism and collective guarantee in the authorities);
• • subjects pursue not only the goals assigned to them by the hierarchy but also other goals often more diligently (e.g., if a corporation bribes power structures to lobby its interests, society suffers from the violation of the mechanism of market competition; if students believe that their goal is to get good grades, not knowledge, then general cheating begins, preventing the realization of the goal of education);
• • in many social hierarchies, the upper levels tend to maximize centralization of governance. The excessive control of the center over all elements of the hierarchy leads to the fact that each of them is deprived of the opportunity to perform the functions of their own maintenance, as a result of which the entire system may perish (history gives many examples of this).

However, hierarchy is not the only possible structure of complex systems. When a system becomes more complex, the distribution of functions across its components can be less asymmetric and centralized, and the importance of parts for the system as a whole becomes less diverse. In such cases, multidimensional and network structures are formed.

This is especially true of social systems in which the essential elements are people with their own subjective goals, which is contradictory to the requirements of the ideal hierarchy. The network is qualitatively different from a hierarchical organization: there is no part of it whose removal would disrupt the functioning of the entire system. For example, the destruction of a direct telephone line between two cities does not prevent them from communicating through a third city to which they are both connected.

Network structures in recent years have attracted increasing attention due to their increasing importance in public practice. Typical examples are marketing and logistics networks in the economy, collective information structures (social networks) on the Internet, distributed computing structures in computer science, and some mass phenomena in sociology and psychology. R. Ackoff [4] drew attention to the fact that the network nature of international terrorism makes dealing with them ineffective, based on the assumption that terrorists are members of an organization and not a network. Networks require a different approach than organizations. The study of network structures has become an actual problem in the theory and practice of management.

Questions and Tasks

• 1. Show using an example that the concept of property as an attribute of an object is a simplified model of its multilateral relations with other objects.
• 2. List four types of relationships between two events.
• 3. Where are the boundaries of the system?
• 4. Discuss the differences between the structure and the structure model of the system.