A Framework For System Study

The development of any system, and particularly Management Information Systems, begins with the development of a conceptual model of reality. "Reality" is a concept that exists within the context of the decision maker who has problems and for whom the new system will provide solutions to the problems. Consider this perception of "reality."

A wife looks up from reading the morning paper and says to her husband, "John, the article says here that men who don't drink live longer!" John ponders this fact for a moment, then responds, "Serves 'em right!"

The purpose of the model is to describe as precisely as possible what the system must do. Once created, the model can be used for more specific development, focusing on how the system's programs are to accomplish their tasks. Ultimately, the conceptual model defines what inputs will be converted in to what outputs via what procedures using what data.

Conceptual models evolve from a definition of what the system is, what it does, and how it is to accomplish its tasks. The definition process focuses on four elements:

Who ownes the system? Who will pay for it? Who will champion its continued existence? For whom is it providing information? Why is the system being developed? Systems that don't have top management support are typically doomed to failure. The owner must be able to provide sufficient resources to ensure that the system is successful.
Defining who owns the system also defines the environment in which the system will exist. As we have discussed previously, systems exist to support the requirements of the environment in which they exist. The environment provides the reason, the purpose for the system. Systems without purpose have no reason for existence, so the specific reason for the system must be defined in terms of its environment.
Who are the users of the system? Who are the principals involved? The system must be usable, or it will not be used. System quality, as defined by Lucas and others, directly influence the users' perceptions of the system. The system must be developed with the ultimate users in mind.
What specific transformations take place? How are the inputs transformed into outputs? What procedures, rules, regulations (government and otherwise), and requirements define and determine the transformation process? What happens as inputs are transformed into outputs?

Defining these four elements provides a conceptual framework for the system's development. By carefully and specifically defining these four elements, the conceptual model takes on an identity of its own. The conceptual model itself then provides a framework for subsequent refinement, specifications, and detailed program development. The conceptual model defines what the system is to do. The "what" is then translated into programs that define "how" the system accomplishes its tasks.

The definition of the conceptual system focuses on identifying verbs that describe the transformation activities. These verbs are used to describe what each specific activity within the system accomplishes. For example, in a sales order entry system, the system may first validate the existence of a customer in the company's system. The sequence of activities is also important to specify. An order for a customer cannot be entered until after the customer's information has been established, and the order for a specific item cannot be accomplished until the item is established in the system. The figure below illustrates the conceptual development process using verbs to indicate actions that the system takes.

A Guide for Initiating System Studies

From Checkland, 1998

Checkland provides some sage advice to the systems analyst (you) that should be considered before beginning any system development process: It cannot be overemphasized, however, that the most important attribute of the good systems thinker is his ability to entertain a wide range of possible relevant systems, to take his choices seriously and to model them meticulously, but to do so without owning them psychologically. As the debate initiated by the model/real-world comparison unfolds, as the 'real issues' emerge, the systems thinker must be able cheerfully to abandon his earlier choices of relevant systems and start again. And again ......

This guide provides a series of questions that system developers can ask (of themselves or others). The questions provide a guide toward identifying the relevant and "important" aspects of the system being considered.

1. The study is undertaken in the context of an owner/champion soliciting the efforts of an analyst (you) in defining what the new system will do for the organization, the environment in which it exists. The analyst should begin by answering the following questions:
- Who is the owner/champion? Who will be the primary support provider for the system?
- Why is the owner/champion requesting the system? What is the purpose of the system? What are the owner's aspirations? In what environment will the system exist?
2. Next, the study focuses its attention on what problems the system will address and (hopefully) solve.
- Who has a problem? Who are the decision makers in the environment? Who will use the system's outputs to support their managerial functions?
- What do the decision makers say about the problems? What are their specific information requirements? In what environment do their information requirements exist?
- Why do the decision makers consider the problems they identify to be problems? Why are the decision makers in the situation they are in? Why do they think that the new system will solve their problems?
- What do the decision makers expect the new system to do for them when solving their problems? How will the system address their problems?
- Based on the answers to the above question, what information and/or activities are important to the decision makers who use the system? What do they want to the system to provide and/or do? Identify as many items and/or actions as possible.
- Based on the answers to the two previous questions, develop some specific names for the situations in which the problems exist and for which the decision makers want assistance.
- Develop a series of nouns and verbs to describe the situation in which the problems exist.
- Identify any environmental constraints under which the problem situation exists.
3. Third, focus the study's attention on the possible solution systems that may be developed.
- Who is/are the problem solver(s)? Who will develop the system that will provide solutions to the people with the problems? Who will be responsible for seeing that the system is developed?
- Who else is part of the system development? What other people will play an active role in the system development?
- What resources are available to develop the system?
  - People
  - Physical and Technological Resources
  - Skills and Knowledge Pool
  - Financial Resources
  - Time (probably the most important resource)
- Identify any environmental constraints under which the problem solvers operate.
4. Identify what conditions and/or situations must exist in order for the system to have successfully solved the problem(s) that it addresses. This is probably the most important issue to address. How do we know that the problems have been solved? How do whe know if the system is successful? Who decides when the problems are solved? Who decides how we know that the system is successful?
5. Who decides what happens to the system once it is implemented? Who decides what modifications, additions, updates, and deletions will be made? Who takes us back to the top of this entire study process so that it can be repeated again as new problems arise?

By following this system study guide, analysts (you) will develop a clear, concise, workable definition of the system to be developed with respect to who wants it, for whom it will provide information, who will develop it, how we know when it is complete, and who decides what happens to it in the future. Greater efforts in developing specific answers to these questions will yield greater clarity and understanding when the actual system development (programming) is begun. Studies have repeatedly confirmed that cost and effort savings of the magnitude of 100 fold can be achieved through careful and thorough initial system study and design. Efforts by analysts in this phase of system design are well worth their time and resources.

The next step in the system development process is to compare the proposed system to "reality." The system's purpose, its anticipated information outputs, and its anticipated uses are compared to the current situation. The purpose of the comparison between "what is" and "what will be" is to identify the desired changes to the current reality. What will the new system change? Why are these changes desirable? Why do we think the new system will accomplish the desired changes? This step attempts to place the new system within the context of its stated purpose and the new environment in which it will create and exist. Once everyone with a stake in the new system agrees upon the "new" reality, the system can be implemented (as described in the SDLC process).

Source: Checkland, P. System Thinking, System Practice New York: John Wiley & Sons

Return to Chapter 1