Systems Analysis takes a goal-centered approach rather than a technology centered approach or a step-by-step chronological approach.

Goal-centered approach starts with defining the goals and going backwards establishing the performance requirements for the preceding step.

Step-by-step chronological approach moves forward by defining the necessary task to carry after the current task is completed.

1. Attributes:
   • Moves from general to specific.
   • Based on goals and objectives.

2. Strengths:
   • Based on the general assumptions and trends of the planning environment.
   • Identifies changes in the planning environment and adjusts to them.

3. Weaknesses:
   • Can lose focus and objectivity if not executed properly.
   • Does not provide sufficient detail for effective action on short-range issues.

1. Attributes:
   • Based on current conditions.
   • Employs current technology plus minor extrapolations.

2. Strengths:
   • Lends itself to immediate evaluation of cost effectiveness.
   • Consistent with conventional engineering design methodology.

3. Weaknesses:
   • Tends to focus vision on short term problems.
   • Produces ever-decreasing incremental improvements.
   • Locks in current technology and operational structure.

1. Systems Analysis is a combination of both top-down and bottom-up approaches as follows:
   • First top-down then bottom-up and alternating between the two.
   • About 80% of the planning is bottom-up and the remaining 20% is top-down.

2. As a result of a SA study:
   • there is a definition of the current state, namely the descriptive scenario.
   • a definition of the preferred future, namely the normative scenario.
   • a definition of how to get from the current state to the preferred future, namely the transition scenario which is action oriented.

• The index of performance concept is either trivial or impractically complex !

• It is the most difficult, unfamiliar, and tension-producing phase in SA.
• It is the phase to which the ultimate success of the system is most sensitive.

THE SEVEN STEPS IN GOAL DEVELOPMENT
1. Generalize the question.
2. Develop a Descriptive Scenario.
3. Develop a Normative Scenario.
4. Develop the Axiological Component.
5. Prepare an Objectives Tree.
6. Validate.
7. Iterate.

1. Generalize the question
   • Helps the client understand the problem.
   • Aid to correct problem statement
   • Helps provide contextual integrity
   • Not a way to avoid solving the client's real problem!!
   1. Why Step 1?
      • The system analyst generalizes the question because he knows the client seldom, if ever, understands his own problem.
      • The outcome of proper generalization is a well defined, quantitative model of the client's "matured" problem.
      • The "matured" problem should include consideration of the impact of the proposed system on non-users and each party involved in the system.
      • By "matured" problem, we mean the problem statement accepted by the client after it has been refined and carefully stated by system analyst.

   2. How far can you generalize? (meaning extend, expand or broaden the question)

      If you generalize too much:

      • It will be difficult to solve it in the time allowed with the money available.
      • The client may get confused, frustrated by the impression that the system analyst is trying to avoid client's specific problem. (Always keep the client happy).

      If you don't generalize enough:

      • You may end up doing exactly what the client told you to do but still not solve his or her problem.

2. The Descriptive Scenario
   • Words or phrases that describe the current condition of the problem and problem environment.
   • Background section, describes current status, state of the art, etc.
   • Use one of the methods of idea generation.
     
   • Brain storming is good
   • Brain writing is better

3. The Normative Scenario
   • The preferred scenario, or preferred future.
   • Start by reversing the descriptive scenario.
   • Extend it.
   • Make sure you work within the client's value system.

4. The Axiological Component
   • The Normative Scenario implies an underlying set of values upon which it depends.
   • Related to or pertaining to human beliefs
   • The conflict in latent values is a major source of grief in public systems
   • The team's values will probably have been intermixed with the client's values.
   • Difficult to discuss client's values: Must be indirect
   • We do not suggest that the system analyst conduct a discussion with the client concerning his or her values because:
   • Many people find it difficult to discuss personal values.
   • To talk about values may seem to some people as attacking them.
   • If the analyst places too great an emphasis on questions of value early in his relationship with the client, he will risk being thought impractical and possibly incompetent.

5. Prepare an Objectives Tree
   • A graphical display of the goals of the system.
   • It helps analysts and decision makers to clarify and organize a rational set of goals in the early stages.
   • It helps in reporting progress and in maintaining goal-directed effort while discouraging non-directed excursions.
   • Constructing the Objectives Tree will initiate thinking about interactions of activities and the relation of objectives to value structure.
     1. Why Manual Construction?
        • Serves to stimulate team discussion/understanding
        • Computer algorithm damps enthusiasm and creativity.
        • Aids in client communication
        • Helps in understanding interactions

     2. Statement Format for Objectives:
        • To (action word) + (object) + (Qualifying words)
        • To (Improve) (IIASA) (Image)
        • To (Improve) (IIASA) (performance)
        • To (Generate) (Positive) (Publicity)

     3. Why this format?
        • Because, it helps keep the objectives simple, clear, direct and action oriented.

        • Goals that display biases of specific team members are accepted initially.
        • They should be eliminated in successive iterations, but to forbid them initially may cause good suggestions to be blocked.

     4. Linkage Semantics
        • Arrows pointing upwards between boxes: "will assist in".
        • Arrows pointing downwards between boxes: "includes" or "implies".
        • We seek a hierarchic ranking of the goals with higher and more general goals above and more specific ones.
        • The goals tree should be expanded until the more specific goals can be measured objectively.

     5. Four Tests of Objectives Tree Logic
        1. Reading down the branch, each goal statement must answer how for its immediate higher goal.
        2. Reading up any branch, each higher statement answers why the goal below is needed.
        3. Reading across the goals at a given level under any one general goal, ask, are all these more specific goals needed to accomplish the more general goal?
        5. Reading across the goals at a given level under any one general goal, ask, what other specific goals at this level are needed to accomplish the more general objective?

     6. Who owns the goal?

        Be aware that different parties involved can have different goals and they may not necessarily come together to form a single logical objectives tree.