LIM, MARYLYN GRACE C.

OOA

            Book: Systems Analysis and Design: an Active Approach

            Author: George M. Marakas

            Reference No.: QA

                                    76.9

                                    S88

                                    M37

                                    2001

Chapter 4: System Requirements Determination

Quote: “Truth comes out of error more readily than out of confusion”

Review:

            In developing an analysis strategy, there are two approaches being use in viewing a system. The first one is called object-oriented perspective. In this approach, the system is being viewed as a collection of objects that interest with each other. The second approach is the process-oriented perspective. This approach tells us that a system can be viewed as a collection of process performed by people in an automated fashion by computers.

            There are three “I”’s in requirements determination. These are impertinence, impartiality and insight. The first of the three “I” is impertinence, this characteristics says that an analyst should question everything and take nothing for granted. The second one is the impartiality; it says that an analyst should be able to consider all issues raised by the system stakeholders to provide the best solution for the organization. And the last characteristic is insight. The concept of insight suggests three things to remember. 1.) An analyst must make the assumption that anything is possible. 2.) They should pay constant attention to detail during requirements determination. 3.) Analyst should think creatively during requirements determination. During the requirement determination, analyst sometimes does commit mistakes. There are four common mistakes in requirements determination. The first common mistake in determining requirement is assuming that the end users of a system are aware of all of their information requirements. Second is collecting requirements from each end user instead of all end users. The third common mistake made by analyst is that they ask the wrong questions so they get less useful responses. The last common mistake is failing to take advantage of the benefits associated with trial and error.

            In requirement determination deliverables, there are six good requirement characteristics that should be followed. First, testable and verifiable, this suggests that requirement must be written such that they represent the acceptance criteria for the new system. Second, justifiable, requirement should be necessary rather than simply desirable. Third, unambiguous, requirement should be stated such that multiple interpretations are excluded. Fourth, modifiable, requirement should allow changes in the business environment. Fifth, consistent, requirement should not be in conflict with any other stated requirement. Lastly, hierarchically traceable, requirement should contain a single system attribute and should be traceable back to a higher level requirement.

            There are two categories for requirement determination techniques that have been identified based on their relative recency of adaptation as information-gathering approaches. The two categories are traditional method and modern method.  In the traditional method, there are four types of information gathering methods. 1.) Direct Interview is meeting with individuals to ask questions about their roles, responsibilities and needs for the current and proposed systems. 2.) Questionnaires and surveys; the analyst submits written, structured question to selected individuals to gather information. 3.) Direct observation is observing individual, process and events to determine the facts surrounding a particular process within a business environment. 4.) Archival document analysis is reviewing recorded organizational documents. The second category in information-gathering approaches is the modern method for requirement determination. There are two kinds of approaches use. The first is JAD. JAD’s main objective is to bring the stakeholders of a proposed system together and use that gathering to facilitate and expedite the collection and determination and consensus on requirements for a new system. The second approach is the iterative prototyping. This helps analyst to quickly convert the basic requirement of a system into a limited working model that can be viewed and tested by the end users.