Marilyn A. Bosito                                    07/01/08 0A                                     OOA

SYANAL       

Mr. Pajo

 

Book: System Analysis and Design with Modern Methods

Author: Len Fertuck

Reference: LRC Extension (QA 402 F471995)

Book Review # 6

Chapter 3

 

“First comes thought; then organization of that thought, into ideas and plans; then transformation of those plans into reality. The beginning, as you will observe, is in your imagination.”  

Napoleon Hill quotes (American author, 1883-1970)

 

 

This chapter shows how to model activities of an organization and group them to form stable subsystem. In this chapter you may learn how to classify kinds of information used at different levels of an organization, identify activities and structure them into hierarchy, identifies which entities uses, compute measures of similarity to identify activities belong together, cluster activities into subsystem of high similarity, and identify groups of activities which constitute a development project. In this chapter also explains how to model activities at the high level of aggregation and how to relate activities to the data they process.

 

An organization is a group of individuals engage in performing coordinated activities. The coordination provide by a hierarchy of formal responsibility, a network of formal relationship, and a network of information. A method of identifying development projects is needed that is less sensitive to this continually changing organizational structure.  An Enterprise model consists of a model of the data used by the organization and a model of activities which enter, use, or transform those data. This information is used to develop a system model that organizes the activities in to groups of related activities. The modeling processes consist of:

·        Structuring activities in the hierarchy.

·        Identifying activities in each level.

·        Relating activities into a data.

·        Grouping activities into subsystem.

Hierarchical structures are natural ways of simplifying problems. A hierarchy permits a complicated structured to be expressed as a set of successively detailed levels. The highest level provides a very general summary of the situation. Each successive lower level reveals mere details about the previous levels.

Activity analysis is a to-down process that identifies aggregate activities that manage resources. Breaks them down into life cycle stages, breaks these into function that must be performed at each stage. Typical resources are money, materials, equipment, personal, suppliers, customers, intangibles, and information. The basic life cycles are planning, acquisition, stewardship, and disposal. Function has identifiable results, identifiable results, identifiable workers, and clear boundaries.  They are self contained, they process information, and they can be automated in single step.

The Business Information Control Study (BICS) Model provides a checklist of activities that an organization must perform depending on the answers to seven questions about transactions. The questions are:

·        Does the supplier bills the customer?

·        Does the supplier deliver the service or product in the future?

·        Does the supplier keeps records about the prior transaction of individual customers for purpose other than billing?

·        Does the customer and supplier negotiate price?

·        Does the supplier rent the product to the customer and retain title?

·        Does the supplier track the product for subsequent result or change?

·        Does the supplier make order?

 

Activities can be related to data entities by using an Activity-Entity Matrix. Each intersection of a row in column indicates whether the activity in the row makes use of data about the entity in the corresponding column. This matrix can be used to cluster activities in to groups that have similar patterns of data use. The results are used for identifying software system and potentially for organization design. The clustering method consists of the following steps:

·        Create an Activity-Entity Matrix

      -This matrix could alternately contain the frequency which each activity use each entity in some based period.

·        Calculate similarities by the matching method for binary data or the fuzzy set method for frequencies.

·        Cluster the activities using farthest neighbor hierarchical clustering.

·        Divide the cluster activities in a groups

·        Check the clustered for balanced