For Scientific Research
One of several approaches to systems development is the SDLC, also called the "Waterfall" model.
| Phase | Activities | Artifacts^1 |
|---|---|---|
| Planning^2 | Define purpose, scope, options, costs^3 | Go/No-Go Decision, Plan |
| Systems Analysis | Process Modeling, Requirements Anal. | Diagrams^4, Requirements |
| Systems Design | Interface & Data Modeling | Diagrams^5, DD^6, Schema |
| Development | Acquisition and/or Coding, Integration | Working System |
| Testing | Testing, Bug Reports & Fixes | Validated System |
| Deployment | Installation, Training, Launch | Production System |
| Maintenance | Backups, Upgrades | Maintained System |
| Evaluation | Effectiveness^7 and Cost Checks | Evaluation Report |
| Disposal | System Removal, Archival, and Surplus | Archived Artifacts |
(See the next slide for notes.)
| Notes | |
|---|---|
| 1 | There are dozens of potential artifacts from each phase. |
| 2 | The planning phase may also be called the preliminary investigation, system investigation, feasibility study, or project initiation phase, among other names. |
| 3 | Determination of options and costs (in the planning phase) is often referred to, specifically, as a feasibility study. |
| 4 | Systems analysis may produce flow charts, swim lanes, data flow diagrams (DFD), and UML diagrams. |
| 5 | Systems design may produce Entity-relationship (ER) diagrams and UML diagrams. |
| 6 | As a systems design artifact, DD represents a Data Dictionary document. |
| 7 | The Evaluation phase will have numerous measures of effectiveness. |
Let's imagine that we are to conduct a research study like this:
- We will sample various surfaces for bacteria.
- Some of those surfaces will be on human subjects.
- We will analyze the bacteria samples in the lab.
- Test results will be entered into a data system.
- We will perform statistical analyses on the test results.
The stakeholder goals for system, in "verb noun" form are:
| Role | Goal |
|---|---|
| Subject | Sign Consent Form |
| Sampler | Collect Consent Form |
| Subject | Complete Subject Form |
| Sampler | Collect Subject Form |
| Sampler | Collect Sample |
| Sampler | Deliver Sample |
| Lab Tech | Test Sample |
| Lab Tech | Dispose Sample |
| Statistician | Analyze Results |
From this goals list, we can create use case diagrams and a structured narrative. A structured narrative is a description which conforms to some sort of standardized format.
For the purposes of our preliminary investigation, we will just add one-sentence descriptions for each goal in our list. Our description will use terms such as "if", "must" or "may" and phrases like "for each" or "for all".
This simple structure will allow us to develop a process model using various types of diagrams. From our process model, we will develop a data model.
| Actor | Use Case | Description |
|---|---|---|
| Subject | Sign Consent Form | If a human subject is sampled, the subject must sign a consent form. |
| Sampler | Collect Consent Form | The sampler must collect the consent form before sampling the subject. |
| Subject | Complete Subject Data Form | The subject must complete a subject data form by providing personal information. |
| Sampler | Collect Subject Data Form | The sampler must collect the subject data form before sampling the subject. |
| Actor | Use Case | Description |
|---|---|---|
| Sampler | Enter Subject Data | The sampler must enter subject information into the system for each human subject sampled. |
| Sampler | Collect Sample | The sampler may collect a sample from a human subject or from a non-human surface. |
| Sampler | Record Geo Location | The sampler must record the geographical sampling location for all samples taken. |
| Sampler | Record Body Location | The sampler must record the human body sampling location for all samples taken from human subjects. |
| Sampler | Record Temperature | The sampler must record a temperature reading for all samples taken. |
| Sampler | Record Humidity | The sampler must record a humidity reading for all samples taken. |
| Sampler | Enter Sample Data | The sampler must enter sample information into the system for each sample collected. |
| Sampler | Deliver Sample | The sampler must deliver the sample to the laboratory. |
| Actor | Use Case | Description |
|---|---|---|
| Lab Tech | Receive Sample | The laboratory technician must receive the sample when it is delivered. |
| Lab Tech | Store Sample | The laboratory technician must store the sample in a cooler (refrigerator). |
| Lab Tech | Split Sample | The laboratory technician must split the sample before testing. |
| Lab Tech | Test Sample | The laboratory technician must test the sample before it expires. |
| Lab Tech | Enter Test Result | The laboratory technician must enter the test result into the system. |
| Lab Tech | Dispose Sample | The laboratory technician must dispose of the sample after it is no longer needed. |
| Statistician | Query Results | The statistician may query the system for test results. |
| Statistician | Analyze Results | The statistician may analyze the results using statistical tools. |
For even more clarity, one can also develop use case narratives, which go into far greater detail. This level of detail might be useful in later design phases.
This diagram shows the actors and their project goals.
This diagram shows a more detailed view of the actors and their project goals.
To better elucidate process branching in a system, we can use a flow chart. Here is a closer look at the sampling process.
This swim lane diagram shows an overview of the system. The lanes not only show actors, but also sequence and outputs.
Context Diagrams (Level 0)
- Context Diagram with Gane/Sarson Symbols
- Context Diagram with Yourdon/DeMarco Symbols and Curved Lines
- Context Diagram with Yourdon/DeMarco Symbols without Subject Entity
Data Flow Diagrams (Level 1 and higher)
- Level 1 Data Flow Diagram
- Level 1 Data Flow Diagram with Curved Lines
- Level 1 Data Flow Diagram without Subject Entity
Working as a group, with a pen or pencil and paper, make some systems analysis diagrams. Sticky notes, a whiteboard, etc. may be helpful when brainstorming and designing the diagrams.
Data Flow Diagrams
Transform your use case diagram(s) into a context diagram and level 1 DFD, especially if there are many data processing steps or data sources. Diagram further levels (time permitting).
Flow Charts and Swim Lanes
Alternatively, if there are complex decisions, branching, or multiple entities performing various actions, then make a flow chart or swim lane diagram from your use case diagram(s).
Explain your diagrams to the group as though you were presenting your data system requirements to those who would build the data system. Ask questions that a system implementer might ask, especially where there are ambiguities.
- Entity-Relationship Diagrams
- MySQL Workbench
- Building Database Tables
- Database Applications
|
|
|
|
 |
|
- Rapid application development
- Creately - Ultimate Flowchart Guide
- Traditional vs Modern Flowcharts
- DFD over Flowcharts PDF
- DFD Slideshow
- Creately DFD
- DFDs - and follow link to "Article"
- UML Data Modeling Video
- Graphical Data Flow Programming in LabVIEW Video
![© Nevit Dilmen](http://commons.wikimedia.org/wiki/File:Happy_Question.svg"){kind=link}
Essentially, all models are wrong, but some are useful.
Source: Empirical Model-Building and Response Surfaces (1987), p. 424. (Wikiquote.)
========================================================
