Systems Analysis
Definition of Systems
There are many different definitions about systems developed over time. Table 3.1 lists three such definitions. The first two definitions seem to place more emphasis on the elements (activities) and do not emphasize the individual's perceptions. The third definition broadens the definition to allow systems to exist that are not just activities but concepts, problems, or an organization.
_________________________________________________________________
Table 3.1 -- Definitions of Systems
System is a set of interrelated elements each of which is related directly or indirectly to every other element, and no subset of which is unrelated to any other subset (Ackoff & Emery, 1972).
System in a set of objects together with relationships between the objects and between their attributes connected or related to each other and to their environment in such a manner as to form an entirety or whole (Schoderbek, Kefalas & Schoderbek, 1975).
System is a mental construct of parts and relationships which make up a whole (Weinberg, 1975).
__________________________________________________________________
Why do we care about the definition of systems? In this course, we will apply systems thinking to "real-world" analysis. Systems analysis developed from scientific thinking which started during the scientific revolution of the 17th century. It is from this period that we have developed the habit of analysis, of breaking down complexity as if this is possible without changing that which is analyzed (Checkland, 1987). In the 17th century, Rene Descartes proposed four rules for using the mind: his second rule urges that, when faced with complexity, the best approach is to split it up into several parts and tackle the problems one by one.
History and Practice of System Analysis
In a sense, system analysis is as old as formal mathematics. However, the application that probably was critical in the formal development of systems theory was commercial telephony. In the 1920s and 1930s, long copper wires were strung around the country, elaborate switching systems were set up, and Bell Laboratories was already at work on theoretical analyses of both engineering and economic considerations about telephone communication systems.
During the 1930s, both radio and industrial electronics were growing in popularity. The first glimmerings of feedback control were evident, and new control devices appeared in the form of synchros and servos. In the 1940s, the Second World War placed accelerated emphasis on technological development of all kinds, including radar theory and development, servo theory and development, and analog computers, the first of which was the Bush differential analyzer. The electronics for all of these devices involved large vacuum tubes which consumed much energy and were not very reliable. The MIT Radiation Laboratory series published in the late 1940s was among the first set of widely available books on system theory and application. At about the same time, the community of theoretical physicists turned to practical analysis of military operations and called it "operations research." The first widely accepted volume on this subject was Methods of Operations Research by Morse and Kimball (1951).
Shortly after the war ended, digital computers began to emerge, and along with them the systems theory appropriate to them. In 1948, Norbert Wiener wrote his widely acclaimed Cybernetics, pointing to the close relationship between electronic systems and animal physiology. In 1949, C.E. Shannon published Communication in the Presence of Noise (summarized in more readable form in Shannon & Weaver, 1963), outlining his theory of information, and not too long afterwards many psychologists were trying to measure how many bits per second could be transmitted to or from humans by various means. By 1950, a number of texts on liner control theory had emerged, and it could be said that by then system theory was alive and well, at least in the world of electrical engineering. The early work of Birmingham and Taylor (1954) and others led to consideration of the human operator as a control system element.
From: Sheridan, T.B. (1988). The system perspective. In Wiener & Nagel (Eds.), Human factors in aviation (pp. 27-51). New York: Academic.
The system approach has proven useful in a number of ways. It is more than a way of drawing diagrams or of doing mathematics. Instead, it is a way of thinking intensively and comprehensively about problems and their interpretations. In this chapter, we will discuss the techniques in systems analysis and how they can be applied to every communication and production procedure. In other chapters, we will discuss Management Information Systems and Systems Engineering.
Systems Analysis Procedures
Systems analysis has been applied to a wide range of production activities and tasks. Since this course is interested in the documentation and communication channels of industry, we will limit our discussion of systems analysis to communication and documentation problems.
Business process analysis recognizes that all business activities, not just production, have quality objectives, and can benefit from standard quality disciplines. Therefore, each and every business process (taking an order in a company, routing an engineering change order, applying for a new job, etc) can benefit from this process. There are six steps to the business process analysis. Each of these steps will be discussed in turn. In Appendix A, there are six examples of business process analyses. I will use the Ten Speed Drive Imports Professional Bicycle Ordering (see flowchart, Figure 3.3a & 3.3b) as an example for discussion.
1. Problem identification
There must be a clear identification and definition of the problem to be studied included scope (from one point to another point in time) of the process and overall criteria of effectiveness, efficiency, and adaptability. In the example, the original problem was to study "the processes involved in ordering a bicycle from Ten Speed Drive Imports (TSDI)."
The term "Feasibility study" implies a study of the practicality of a proposed project. It involves a preliminary analysis of the total requirements needed for the evaluation of the problem. Some problems might be easily defined but also might be too large to study adequately. In addition, it might not be possible to obtain all required information about a problem or there might be legal problems concerned with the analysis of the problem. If it were not possible to obtain the forms from TSDI, this problem would have been valid, although unfeasible.
3. System definition (detailed study)
The purpose of this phase is to obtain a clear, concise and bounded definition. The subject of the proposal and its limits, and well as the objectives and anticipated benefits, should be specified. This section covers a wide range of activities: (1) investigation of the current (existing) system, and (2) analysis of the results, and (3) identification of the problems in the current system. The investigation of the current system is your assignment for project 1 (see handout). This section will be discussed further in this chapter.
4. Systems Design
The purpose of this phase is to suggest changes to transform the current system by altering the performance and activities to solve the problems that were identified in section 3. The discussion of problems and their solutions is your assignment for project 2 (see handout).
5. System Implementation
This section is where you implement the solution(s) you found in section 4.
6. System Evaluation
This section is included to reevaluate the effectiveness, efficiency , and adaptability of the changes that were made to the process.
The above descriptions will fit any business process analysis; however, I would like to give more specific details on how to proceed.
Define the problem. " I am going the study the registration process at SJSU in the history department from the point I buy the semester's schedule of classes until I successfully register for classes using the computerized registration system." Note that there are three parts to this problem: A specific, clear problem; a starting point; and an ending point.
Flowchart the process. In order to study the problem, I need to break it down into major activities (seeing my advisor, getting the course catalog, checking my work schedule, etc). After listing all the activities, I need to flowchart the major activities. The symbols used for designing flowcharts are shown in Figure 3.2. These symbols are identical to those used in Computer Science and they have similar meanings. The rectangle is used for activities and the diamond is used for decisions. A process step (rectangle) can have many inputs (lines coming in) but only one output. In contrast, A decision (diamond) requires multiple outputs. When using the decision block, you are asking a question. Questions necessarily have more than one answer. Figures 3.3a and 3.3b show a flowchart that shows the processes involved in ordering a custom-made, ten-speed bike.
Define inputs and outputs. For each activity, define inputs and who supplies them and outputs, and who receives them. Back to my registration problem, after I buy the Schedule of Classes, I scheduled an appointment to see my advisor. The process step (rectangle) that including buying a Schedule of Classes was the input for the process step (rectangle) for scheduling an appointment with my advisor. The output of scheduling an appointment with my advisor was the input to another process step.
Analysis of the Current system. After designing a flowchart and analyzing the current flow, you should be able to identify problems or roadblocks with the process as it exists. You should identify the problems in clear and concise terms. For example, in the registration problem, it was difficult to schedule an appointment with my advisor.
Identify solutions and alternatives. Every problem can lend itself to many different solutions: some more feasible than others. You should attempt to find three solutions to each problem that you have identified. For each solution, you should define any changes to the current process and analyze the different solutions in terms of feasibility and cost. Identify the best solution for each problem and justify why it is the most feasible. (For more details on this section, see handout on Project#2).
Implementation Plan. An implementation plan should be included in each Project 2 analysis. If you undertook this analysis in a real-world situation, it would be submitted for approval before implementation. After the changes were approved, you would then implement them. Therefore, a projected implementation plan should be included.
Evaluation. After implementing any change to a system, it is necessary to evaluate the effectiveness of the change over time. It will not be possible for you to do this in this class; however, it is one of the steps in undertaking a process analysis.
Additional Resources
In Appendix A, there are copies of different Projects 1 and 2. I attempted to include as wide a range of problems that have been studied. Any business or manufacturing process can benefit from a detailed, step-by-step analysis. In this class, we are particularly interested in investigating processes that requires documentation or communication.