Sunday, February 26, 2012

Evaluating Data Flow Diagram Quality

In the last two assignments, we were told to develop an activity diagram and a fully developed explanation for a use case and generate at least 3 different styles of Data flow diagram of USEP's pre-enrollment system, respectively. As part of our next assignment, we were told (as an analyst) to define the characteristics does an analyst examine when evaluating DFD quality.

Before discussing thoroughly what was being asked and since not all of us are familiar with data flow diagram, I will define it first, as well as the different notations incorporated with data flow diagram.

Data flow diagrams are use to illustrate the different processes integrated in a system. It has inputs and outputs. Those inputs undergo a process or different processes to produce an output or outputs. According to Gane and Sarson, data flow diagrams (DFDs) were established and well-liked for Structured Design Analysis. Data flow diagrams give us idea about the flow of data or information: from the exterior component into the system, then demonstrated how the data stimulated from one process to another plus its logical storage.



In creating a data flow diagram, there are only four symbols needed. 

1) SQUARES – it is used to represent the external components or entities, which are the sources (starting point) or the destinations (end point) of data

2) ROUNDED RECTANGLES – it is used to symbolize the processes: having the data as the input, then doing something to it and produce an output from it.

3) ARROWS - which can either be electronic data or physical items

4) OPEN-ENDED RECTANGLES – it is used to represent the data storage. It includes electronic storage such as databases, and physical storage such as filing cabinets.


DATA FLOW DIAGRAM NOTATIONS

Process Notation

A process is the action of taking the incoming data flow into out coming data flow, based on the establish set of steps or procedures.



Data Store Notation

Data stores serves as the warehouse or depository area of data in the system. Sometimes, data stores are called files.




Data Flow Notation

Data flows perform as channels by which packets of information flows. It bridge the gap between a process to another process. Usually, the arrows which represent as data flow channels are labeled with the name of the data that passes through it.





External Entity Notation

External entities are pertaining to the objects located outside the system, through which the system interacts. External entities are, basically, the beginning and ending of the system’s inputs and outputs.




Context Data Flow Diagram

Of all diagrams, context data flow diagram in most basic data flow diagram since it serves as the level 0 diagram. Context free diagram is incorporated by one process node only, which is the process 0, that take a broad view of the function of the whole system in association with external entities.





• The context diagram explains the system limitation or system boundary clearly 
• It does not typically exhibit data stores because all of the system’s data stores are reflected on within the system capacity 
• Context diagrams are frequently generate  at the same time with the event table.


In the context data flow diagram, the analyst will prepare the first level of the data flow diagram which will illustrate the whole flow of information in a system in a single process. Basically, the context data flow diagram will show how the analyst view the whole information system, including the user/s and the interacting actors. For instance, in our previous assignment, the student and the cashier for the payment process. Most of the time, the context data flow diagram does not have any data stores. The context data flow diagram only shows the major data inflows and outflows.


Physical Data Flow Diagram

A physical data flow diagram shows the clearer division of events and process as well as the different actors interacting in the system and the required data needed as an input to have the desired output. Also, the physical data flow diagram differentiate the manual and automated procedures as illustrated in its diagram.

In the physical data flow diagram, the analyst must be able to define the different processes in detailed, must identify the possible and temporary data stores, as well as to be able to know when and where to add controls in order to have more effective yet simple system. 

Usually, physical data flow diagrams are established and utilized at some point in the final stages of analysis or early phases of design. Physical data flow diagrams are very helpful in terms of defining the alternate execution of a system antecedent to a more comprehensive deign models development. Analysts should always be aware on the possible effect if he will do the creation of physical data flow diagram during the analysis phase activities. Analysts must avoid that one except if he is in the midst of producing alternatives. Analysts should not forget to label physical data flow diagrams to be able for the readers to know that the model shows one probable execution of logical system requirements.

Logical Data Flow Diagram

Another type of data flow diagram is the logical data flow diagram. A logical data flow diagram present the flow of the data in the system, how the system operates and how the actor involved in the system process interact with one another. The logical data flow diagram defines the different procedures incorporated in the information system, the data that is required (as to be the input) as well as the possible output after undergoing an event or process. Also, the logical data flow diagram does not necessary go into details in the physical implementation of events when illustrating the activities involved in logical data flow diagram.

In the logical data flow diagram, the analyst must ensure that it will help the employees in the organization communicate easily. Also, the analyst must illustrate a high quality logical data flow diagram which will attend to produce a more established and secured system.The analyst needs to be flexible and needs to have a thorough understanding in the logical process of the system to be able to maintain the system and to help the user to easily abolish the redundancies that could be found in the system. 




As a systems analyst, I must make certain that the data flow diagram I am working at is readable, has a consistency in internal entities, ensure that the system requirements are presented precisely and make sure that it has diminished complexity. To ensure the high quality set of data flow diagrams, analyst can apply the a few simple rules in making data flow diagrams. These rules can be applied by the analyst either in these ways: while developing the data flow diagram, or during a separate quality check after preparing data flow diagram drafts.


We all know that, as human, we have limited capability of analyzing and understanding complicated information. Information overload is a term used when a great extent of information is presented all at once. As a good systems analyst, one must divide that great amount of information into subsets to avoid information overload. Dividing this large set of information with the use of data flow diagram would help, not just the analyst, but also those people who are interested to acquire those information, to understand and analyze the information clearly.


As a system analyst, he or she is capable in detecting inaccuracies and exemption in a set of data flow diagram by means of looking for discrepancies. There are three usual and easily identifiable consistency errors in making data flow diagrams, according to Satzinger, Jackson, and Burd.

• Differences in data flow content between the process and its process decomposition
• Data outflows without corresponding data inflows
• Data inflows without corresponding data outflows.






Resources:

http://www.smartdraw.com/resources/tutorials/data-flow-diagrams/#/resources/tutorials/Introduction-to-DFD
http://me.emu.edu.tr/ie447/CIMLectureNotes2011.pdf
http://www.differencebetween.com/difference-between-physical-dfd-and-vs-logical-dfd/
http://books.google.com.ph/books?id=onoxYRropMoC&pg=PA234&lpg=PA234&dq=what+characteristics+does+an+analyst+examine+when+evaluating+DFD+quality&source=bl&ots=Xiov4kUceS&sig=ePlTUEmtMIw1PEW-EoHYY-Of5Xc&hl=en&sa=X&ei=2mo0T-K0POHemAWa8KGLAg&ved=0CBwQ6AEwAA#v=onepage&q=what%20characteristics%20does%20an%20analyst%20examine%20when%20evaluating%20DFD%20quality&f=false

No comments:

Post a Comment