Rapid Prototyping

Rapid Prototyping is divided into these main sub phases:

• Develop a Prototype.
• Test the Prototype.
• Gather more Requirements.

Knowledge Engineer develops a Prototype based on the detailed requirements or knowledge acquired during the analysis and design phase. Prototype is the demo version of the actual expert system (ES). Prototype is then tested. Knowledge Engineer performs Alpha Test to gather more requirements. He calls the customer/s at his site to execute the prototype and asked him/them to comment on the prototype.

System Development

System Development has two main sub phases:

1. Develop the actual ES.
2. Evolve the ES.

There are two main approaches to develop the ES.

1. Throwaway Prototyping.
2. Evolutionary Prototyping.

Throwaway Prototyping approach is used when the knowledge engineer throws every version of the prototype after the approval of the customer. He develops the ES software totally from the scratch. Evolutionary Prototyping is used when the knowledge engineer used the prototype to develop the ES software and evolves it continuously until customer approved it.

Implementation

These are the main implementation phases:

• Acceptance Testing.
• Beta Testing.
• Installation.
• Integration Testing.
• Training.
• Security.
• Documentation.

Knowledge engineer conducts Acceptance Test by executing the ES software in front of all users. It is very important that users accept the interfaces and functionalities of ES. Most of software in the real life has been failed because users refused to accept them. Beta Test is conducted at the customer site before installation of the actual software. ES is tested with real test data in the environment where it is going to operate. ES software is then installed at the customer site. ES is tested for its integration with other information systems.

Post Implementation

Post implementation phase has three main sub phases:

1. Start Operating.
2. Maintenance.
3. Evaluation.

ES starts operating in the field. Maintenance is a process to remove errors in the software. It starts immediately after developing the first prototype of the ES. Knowledge Engineer evaluates the ES time to time to check its performance, efficiency and accuracy. The main objective of performing the evaluation phase is to compare the maintenance cost against the benefits achieved after implementing the ES. ES usage is stopped if maintenance costs exceeds the costs saved due to implementation of ES, such as traveling, salary and overtime charges costs.

Small Scale Expert System SDLC

Small Scale Expert System SDLC has five main phases:

1. Identification.
2. Conceptualization.
3. Formulation.
4. Implementation.
5. Testing.

Identification phase is covering two main phases of Large Scale Expert System SDLC:

1. Project Identification and Selection.
2. Project Initialization.

Knowledge Engineer tries to conceptualize the ES with respect to problem domain during the conceptualization phase, such as what is the information required to develop the ES and how it can be represented.

Formulation phase is used to gather the knowledge from the main sources of knowledge. Knowledge Engineer made a summary of requirements after performing the knowledge acquisition process.

A Prototype is developed based on the requirements during the implementation phase. Prototype is tested to develop the actual ES software following the throwaway or evolutionary approach.

Related posts:

1. Methodologies to Develop Expert Systems (Part One) These are the two main methodologies used to develop expert...
2. Methodologies to Develop Expert Systems (Part Two) System Analysis and Design Main phases of System Analysis and...
3. Main Purposes of Developing Expert Systems and Main Concepts of an Expert System (Part One) Main Purposes of Developing Expert Systems The main purposes of...
4. Prototype and Types of Prototype and Rapid Application Development Prototype Listen to the customer. Develop a demonstrating (demo) version...
5. Components of an Expert System These are the main components of an expert system: Knowledge...