"Systems Analysis and the Systems Approach" by S.P. Nikanorov
"Systems analysis" is understood as a methodology for solving complex problems in the development of industry, transportation, defense, education, and other domains, as well as problems of organizational design.
Systems analysis was initially applied for selecting an individual product from a set of functionally similar products. However, it later became clear that such selection could only be carried out within the framework of broader goals, where the products under consideration were merely subsystems and merely stages of development. Thus, increasingly large parts of a production or military system became objects of study. The methodology of systems analysis gradually became the foundation of a continuously operating mechanism for solving a large set of interrelated problems. The most important function of systems analysis consisted of determining the structure of connections between tasks and the possibilities for their solution. The structures thus discovered served as the basis for selecting goals and the means of achieving them. The methodological instrument of systems analysis — its conceptual schema — is primarily designed to perform this function. The practical experience expressed in the conceptual schema, once mastered, becomes the foundation for an individual's or a team's problem-solving actions.
The following reasoning helps explain the factors that determined the conceptual schema of systems analysis. At the center of this methodology lies the operation of quantitative comparison of decision alternatives. Quantitative assessments must characterize the compared alternatives of equal quality in terms of their effectiveness and the costs of achieving a given result. The correctness of quantitative assessments depends on how fully and accurately everything that affects effectiveness and costs has been taken into account. Thus arises the idea of identifying "all elements related to a given alternative," that is, an idea expressed in natural language as "comprehensive consideration of all circumstances." The complex of elements identified by this definition is what systems analysis calls a "complete system."
But how does one identify this complex of elements, this "system," and how does one establish whether a given element belongs to a given alternative or not? The only criterion can be the participation of a given element in the process leading to the output of a given alternative. Since this is so, the concept of process turns out to be the central concept of systems analysis, around which the entire conceptual schema is built.
A system is defined by specifying system objects, their properties, and the relations among them. System objects are the input, the process, the output, as well as feedback and constraint.
The input is that which precedes the occurrence of the process. One can also say that the input is everything that changes during the process. Participation in the process is established by the presence of change. The input consists of input elements. In some cases, the input elements are the "working input" (that which is "processed") and the processor (that which "processes").
The output is the result or the final state of the process. The process transforms the input into the output. The ability to transform a given input into a specific output is called a property of the given input.
A relation determines the sequencing of processes; that is, the output of some process is the input of some other specific process. Every input of a system is an output of this or another system, and every output is an input. Identifying a system in the real world means specifying its inputs, processes, and outputs. Every system consists of subsystems. Every system is a subsystem of some system. It is postulated that any real-world object can be described in terms of system objects, properties, and the relations among them.
Artificial systems are those whose elements are made by people, that is, are outputs of consciously organized processes. In every artificial system, there exist three subprocesses differing in their role: the main process transforms the input into the output; feedback ensures correspondence between the actual and the desired output by modifying the input; the constraint-handling process ensures correspondence between the output of the system and the requirements for it as an input to the subsequent system that is the consumer of this input.
In the feedback subsystem, a number of operations are performed: a sample of the output is compared with the output model and their qualitative-quantitative difference is identified, the content and meaning of the difference are assessed, a decision arising from the difference is developed, and a process for introducing the decision (an action on the input) is formed. The constraint-handling process is triggered by the consumer of the system's output, who analyzes this output. This process acts on the system's output, accepting or rejecting it, and on the model of the system's output. The output model, reflecting the constraint, defines the goal (function) of the system and the constraining relations (qualities of the function), which, through the constraint-handling process, are harmonized with the goals of the consumer.
A problem situation exists when there is a difference between the necessary (desired) output and the existing output, which may manifest as symptoms. The existing output is provided by the existing system. The desired output is provided by the desired system. The problem is the difference between the existing and the desired system. The problem may be either preventing a decrease in output or increasing output. The condition of the problem represents the known. The problem's requirements represent the desired system. The solution to the problem is a system that fills the "gap" between the existing and the desired systems.
To solve a problem means to construct a system which, together with the modified existing system, constitutes the desired system. The process of finding a solution to a problem centers on iteratively performing operations to identify (elucidate) the condition, the goal, and the possibilities for solving the problem. The result of these operations is a description of the condition, the goal, and the possibilities in terms of system objects, that is, in terms of the structure of relations and subsystems. If the structure of relations and subsystems of the condition, goal, and possibilities of a given problem is known, the identification has the character of determining quantitative relations, and the problem is called quantitative. If the structure of relations and the subsystems of the condition, goal, and possibilities are known only partially, the identification has a qualitative character, and the problem is called qualitative or ill-structured.
Systems analysis establishes the fundamentally necessary nomenclature of functions for problem solving, which consists of: detecting the problem, assessing the urgency of the problem, defining the constraint (the goal and constraining relations), defining criteria for measuring the degree of approximation of the problem's solution to the desired one, analyzing the existing system, determining the structure of possibilities for constructing a set of alternatives, selecting the solution alternative, securing acceptance for the chosen solution, making the decision (assuming formal responsibility), implementing the decision, and determining the results of solving the problem. The boundary of the problem-solving process is determined by the conditions, goal, and possibilities for its implementation.
By postulating the nomenclature of problem-solving functions, systems analysis provides a means for analyzing and designing organizations. In particular, organizations with a hierarchical structure of subordination can be transformed into problem-solving organizations by assigning problem-solving functions to their subdivisions. Such are, in the most general terms, the basic concepts of systems analysis as a methodology for solving problems.