II. Systems Practice: Technology of Applied Systems Analysis
Technology of Applied Systems Analysis
In accordance with our system concepts, set out in Part I, the transition from the state of a problem situation to the state of a desired final goal — the solution of a problem — must be carried out systematically and orderly by consistently performing certain steps. In addition, each stage has its own structure of smaller steps, which should be followed fairly strictly; its violation may adversely affect the quality of the result of one stage and, consequently, of the entire process. In addition, at each stage, there are dangers to make a mistake or fall into a “trap”; you need to know about the possibility of such failing actions and use suitable techniques to avoid them. The technology of applied system analysis is presenting this entire algorithm with a description of all the features of each stage.
Having come to the system analyst with a problem the client could not solve himself, he initiates the systems analysis procedure. The analyst will take up work to solve any problem, but only under certain conditions. These conditions are absolutely necessary (although not sufficient) for success. Without them, experienced analysts simply do not get to work. Let us list these conditions, although their full meaning and necessity will become clear later after getting acquainted with certain points and subtleties of the technology.
Success conditions for system research:
- 1. guaranteed access to any necessary information about the problem situation (while the analyst, for his part, guarantees confidentiality, in case of its necessity);
- 2. guaranteed personal participation of the first persons in organizations, obligatory participants of the problem situation (heads of problem-containing and problem-solving systems);
- 3. the rejection of the requirement to preliminarily formulate the necessary result (“technical task”, “terms of reference”) since there are many improvement interventions that are not known in advance, all the more chosen for implementation.
Operations of Systems Analysis
If the client agrees to the terms of the contract, the analyst proceeds to the first stage, after which he starts the second and so on until the last stage, at the end of which the implemented improving intervention should be obtained (Figure 5.1).
From the properties of the composition model, we know that the whole can be divided into parts differently (the third property of the systems); therefore,
FIGURE 5.1 Technology of solving a simple problem.
different authors offer descriptions from different numbers of stages: descriptions can be enlarged and divided. For example, the path from posing a problem to solving it can be divided into two stages: research and resolution. Each of them can be divided into two stages: examination and diagnosis, planning and implementation of the plan, and so on, in full accordance with the method of analysis (see Chapter 3 in Part I).
We prefer to present a system analysis algorithm in a dozen stages. The scheme of presentation of each stage is standard: its input is given; it is described what and of what quality should be obtained at the output; it is indicated what actions and how should they be performed; it draws attention to possible difficulties, mistakes, and “traps” and gives advice on how to overcome them.
It would be wrong to think that a consistent linear passage of all stages will always lead to the desired solution. This happens only when the contractor knows in advance the complete solution to the client’s problem. For example, when a teacher conducts classes with a student, teaching him how to add fractions, explaining to him every step in the process.
However, in the practice of applied systems analysis, the final result is not known to anyone in advance; it will be gradually formed during the analysis. Therefore, systems analysts refuse to accept from the customer “technical task” at the beginning of the work, which defines w'hat should be the result of their work; therefore, systems analysts do not agree to “justify” someone who has already made decisions. A systems analyst cannot be compared with a doctor who diagnoses and prescribes treatment, and not even with a teacher who teaches what he knows, but rather with the head of the creative team who has taken on a new topic without knowing how it will end.
Consequently, the subsequent stages of the analysis may reveal incompleteness or error of some of the previous stages; it will be necessary to return to it and correct the discovered deficiency, once again going through the stages already passed. The scheme of the analysis procedure is shown in Figure 5.2.
The algorithm for solving a particular problem will not be linear, and it will contain cycles and returns. In fact, this is a different representation of the trial and error method: the solution to the problem is to overcome the complexity (!). The more complex is the problem, the more returns will be required (their number can be considered a measure of complexity of the situation).
However, in the future presentation of the technology, we will discuss the stages sequentially, one after another. The actual, real progress can take the shuttle path until you reach a final decision.
FIGURE 5.2 “Trials and errors” method for solving complex problems.
A common misconception, the main cause of failure, is the opinion that for any problem situation there is a simple and a correct solution. This is true only for cases where the set of problems is connected in a linear sequence, and the problem of the strength of the entire chain is to strengthen the weakest link.
However, such a situation is a rare exception; problems form a system whose emergence does not allow one problem to be solved individually and requires an effort to create an improving intervention (see Chapter 1, Part I).
Finally, it should be emphasized that both the operations of the systems analysis and the specific variant of the progressive-return process according to scheme of Figure 5.2 depend on the paradigms used (vision of the world, the set of ideas and assumptions on which we act) and metaphors (specific ideas about the system under consideration). Therefore, further presentation of the stages of the system analysis will be conducted taking into account different options for their execution.
About Various Options for Solving Problems
Even “hard” problems sometimes have various solutions (e.g., proofs of a mathematical theorem). For “soft” problems, this is a rule, not an exception. This is particularly evident and typical of the problems encountered in the management of social systems, with their diversity, complexity, and poorly predictable variability.
In response to the needs of managers in specific management practices in terms of the complexity of the managed system and the turbulence of the environment, there are various techniques advertised as a panacea for all managerial ills. Here are just a few of them:
- - value-added analysis;
- - core competencies;
- - balance-sheet accounting;
- - total quality management;
- - continuous improvement;
- - optimization of staffing (downsizing, rightsizing);
- - user orientation (customer focus);
- - planning in the form of scenarios (scenario planning);
- - business process re-engineering;
- - self-learning organizations;
- - knowledge management;
- - establishment of criteria and standards (benchmarking).
The practice of using such techniques shows that they rarely give the desired improvement. Although there can be several reasons for frequent failures, there is one fundamental panacea common to all: inability to see the integrity of the system, focus on its individual parts, disregard for the main features of systems, that is, emergence and synergy.
Modern applied systems analysis offers a holistic and creative approach, which focuses on two “system-forming” factors: (1) the integrity, emergence, synergy of the system (the inadmissibility of a separate consideration of any part when the goal is to improve the whole system), and (2) the entry of the system as part of a larger, embracing system, and the interconnectedness of the system with other systems in the environment (the need to take into account the integrity of the encompassing metasystem; consideration of the problem situation from several different viewpoints). In his fundamental survey  of different approaches to solving various problems, M. C. Jackson described in detail 10 such technologies used in situations that differ in a level of disagreement between stakeholders, and in a degree of exactness of the available model of real situation. Each technology is a specific (adjusted to the situation) combination of all the obligatory (for success) operations, described in the following text.
Stage One. Fixation of the Problem and Problem Situation
The task of this stage is to formulate the problem and document it. The problem statement is generated by the client; the role the facilitator is to find out what the client is complaining about and what he is unhappy with. This is the problem of the client as he sees it.
The task of this stage is to formulate the problem and document it.
Thus, it is necessary to try not to influence his opinion and not to distort it. In conversation it is better not to submit a remark like “I agree (disagree) with you”, but only “I’m listening to you”. (This rule should be followed when interviewing at subsequent stages: after all, we need information not from the analyst but from those with whom he talks.)
The most serious mistake at this stage would be to immediately address the problem. This cannot be done for a number of reasons.
- 1. The client asked for help because he could not solve the problem on his own as the situation was difficult for him. Therefore, his model of the situation is inadequate, and he can only report the situation that he knows about it. Therefore, an attempt to solve the problem at once only with this information is doomed to fail. In fact, the next steps are designed to collect missing information to ensure the adequacy of the model.
- 2. Subsequently, it will become clear, including to the client himself, that the initial definition of his problem is inaccurate, or incomplete, or even incorrect. Often times the client diagnoses the problem and at the same time is mistaken. Symptoms are often mistaken for causes.
For example, when a patient comes to the doctor with a complaint of pain in the left hand, the doctor will fix the anamnesis, but will not immediately treat the hand, directing the patient to a cardiogram: perhaps this is an indication of heart disease.
Consider another example: the firm complained that the production departments do not have time to produce parts to customer order, and it turned out that the delay is excessively long in prior clerical and order processing.
3. In the course of the study, it may turn out that to eliminate the problem of the client, it is not necessary to solve it, but someone else’s, a completely different problem.
For example, employees of a large organization complained about the long wait for elevators in their multistoried building. Technical services offered the following options: (1) to build additional elevators, (2) to install high-speed elevators instead of the old slow ones, and (3) to create a single dispatching control of elevators. The psychologist offered to hang large mirrors in the pre-lift halls. The option turned out to be the cheapest and most effective: women engaged in preening, men quietly spied on them, and the complaints stopped.
In another example, to solve a number of student problems, it is necessary to solve the problems of teachers first; to solve the problems of patients, it is necessary to solve the problems of medical staff.
Thus, fixing the client’s problem is only a starting point, the beginning of a systematic study, and not a ready formulation of the problem to be solved immediately.
As for the fixation of the problem, it should be noted that the documentary design of the work done not only at this but also at subsequent stages is necessary in connection with the shortcomings of human memory, increase in the volume of information in the course of work, changes in the environment over time, etc.
Questions and Tasks
- 1. Why is it necessary to document the client’s problem?
- 2. Why not start solving the problem immediately after fixing it?