 |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  | |
| By N2H | ||||||||||||||||||||||
Process and data modeling
October 25, 2007
If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting and have a nice day!
System Analysis is primarily concerned with:
-
Finding facts that will permit the understanding of the present system and aid the design of any successor.
-
Using fact finding techniques that will enable the finding of system requirements.
-
Organizing the facts into a rigorous set of documentation.
In-order to gain proper understanding of these facts, various techniques, tools and methods are used.
Modeling
A model is an abstraction, a representation of part of the real world. It may concern a representation of one aspect of the present or proposed Information system.
Emphasis in Information System has been placed on process modeling and data modeling.
All these modeling techniques help in occupying and design the new/future Information System.
Process Modeling
Although there are many process modeling techniques, all of them have as their unifying elements emphasis on the process and the basic technique of functional decomposition. Some of these techniques associated with structured Analysis & Design and functional decomposition are:
-
Decision trees
-
Decision tables
-
Data flow diagrams
-
Data structure diagrams
-
Structured English.
The techniques and tools of process modeling help in understanding the real world processes and communicating the knowledge acquired.
They communicate these through their tools. Most of them are graphical and this encourages user involvement. Data flow diagramming are particularly useful in communicating the analyst’s understanding of the option.
DFD provides a means of achieving structure in systems; in that it enables a system to be partitioned into smaller or desirable size so that the system can be understood easily.
DFD provides information in a graphical and concise manner. The graphical aspect means that DFD can be used as a static piece of documentation and a communication tool, enabling communication of all levels i.e. Analyst to user, Analyst to programmer, Analyst to Analyst etc
The fact that users can readily understand DFD means that it can be easily validated for corrections and thus increases the chance of a successful Information system.
DFD is concise in that it allows for a system to be examined at the highest level i.e. an overview and still allows it to be viewed in detail, whilst maintaining the links and interfaces at the different levels.
DFD provides the analyst with the ability to specify the system at a logical level, thereby providing independence between the logical and the physical implementation of the system, thereby allowing the users to specify their requirements without restrictions of physical implementation.
A logical DFD represents logical information of what flows into the system whether its customer credits. It does not bother with how it flows e.g. by twisted copper.
Data Modeling:
Data modeling concentrates on understanding and documenting data. Data is considered as the fundamental building blocks of systems. The Data model is a result of data analysis, and it is oriented towards that part of the real world that it represents i.e. org, dept. etc.
The data model should always be implementation independent in that the data model and the data analysis that derives it is suitable, whether the principal model is a database, file card etc.
The success of a data model comes about with the systematic way in which it identifies the data in organizations and the relationships between them i.e. data structures.
Data analysis techniques attempt to identify the data elements and analyze the structure and meaning of data in the organization. This can be achieved by interviewing people in the organization.
Studying documents, observation, questionnaires etc and formalizing the results through a process called Entity Modeling. Graphical documentation aids in the process of data analysis.
Data analysis is used to help in understanding aspects of a complex organization. Good data models can be used as discussion documents for understanding aspects of the organization and the process as well as improving the effectiveness of the role of data and information in an organization.
Data model/analysis is stressed by many methodologies mainly because data is more stable than processes, also because:
-
The data model is not computer-orientated i.e. it is not biased by any particular physical storage structure that may be used because it provides for logical/physical independence i.e. the model stays the same whether the storage structures are held in the magnetic tape, disk or mini storage.
-
The data model is a model that is understandable by everyone i.e. the users, developers etc. This is a very easy to understand it and validate.
-
A data model is able to reflect a variety of different view of data across departmental or section areas.
-
A data model is readily transformable into other models such as relational, hierarchical or network which are mostly used to represent data structures in a DBMS.
-
Available data modeling/analysis techniques allow a choice of alternative methods to be used where appropriate thus one technique can be used to cross check another.
-
Data modeling/analysis is rule based which means the result of one analyst’s work can be followed and proven. Data model can be adapted to appear in different forms for different users it can also appear as a whole
There are various approaches to data analysis; some are the data collection approach (i.e. document driven analysis). The documents used in a department or organization are analyzed in a bottom-up manner. These documents include reports, forms etc.
The analysis of each document will in turn lead to the formation and then improvement, of the data model showing the data and the relationship between these data.
Another approach is entity modeling. This approach gains its information by interviewing people in the organization. Entities such as customers, suppliers and their relationship are ascertained and represented as a graphical presentation
Comments
Got something to say?
You must be logged in to post a comment.
Powered by FireStats