Analysis, Design, and Implementation of an Agent Based

1

• Analysis, Design, and Implementation of an Agent
  Based
• Management Information System
• Supervisor Dr. Razzazi
• Advisor Dr. Rahmati
• By Leila Homaeian
• Summer 1381

2
General Concepts

• Agent Definition
• Ontology
• Agent Characteristics of Java

3
Agent Definition

• Reactive
• Autonomous
• Goal-Driven
• Temporally Continuous
• Orthogonal Properties
• Communicative
• Mobile
• Learning
• Believable

4
Seven Reasons for Using mobile
agents

• Reducing the Network Load
• Overcoming Network Latency
• Encapsulating ?Protocols
• Executing Asynchronously and Autonomously
• Adapting Dynamically
• Naturally Heterogeneous
• Robust and Fault Tolerant

5
Ontology

• What is an Ontology?
• A Simple Knowledge-Engineering Methodology

6
A Simple Knowledge-Engineering
Methodology

• Determining the Domain and Scope of the Ontology
• Considering Reusing Existing Ontologies
• Enumerating Important Terms in the Ontology
• Defining the Classes and the Class Hierarchy
• Defining the Properties of Classes- Slots
• Defining the Facets of the Slots
• Creating Instances

7
Agent Characteristics of Java

• Platform Independent
• Secure Execution
• Dynamic Class Loading
• Multithread Programming
• Object Serialization
• Reflection

8
Project Description

• System Identification
• Methodology for Analysis and Design MESSAGE
• Methodology for Engineering Systems of
  Software AGEnts
• Case Tool MetaEdit
• Tool for Implementation ASDK
• Aglets Software Development Kit

9
System Identification

• Requirements Identification
• Components within the System
• How the Functionality of the System is done?

10
Analysis and Design of the Registrar
office

• Main MESSAGE Concepts
• Analysis Model Views
• Analysis Process
• Analysis Diagrams
• Design Process

11
Main MESSAGE Concepts

• Foundations
• Knowledge-level Concepts

12
Knowledge-level Concepts

• Concrete Entities
• Activities
• Mental State Entities

13
Concrete Entities

• Agent
• Organization
• Role
• Resource

14
Activities

• Task
• Interaction and Interaction Protocol

15
Mental State Entities

• Goal
• Information Entity
• Message

16
Analysis Model Views

• Definition of Views
• Consistency between the Views

17
Definition of Views

• Organization View
• Goal/Task View
• Agent/Role View
• Interaction View
• Domain View

18
Analysis Process

• Introduction
• Inputs to Modelling Process
• Overview of Modelling Process

19
Introduction to Analysis Process

• Refinement Approach
• Analysis Refinement Strategies

20
Components within the system

• Interface Agent
• Student Role
• Administrator Role
• Broker Agent
• Broker Role
• Staff Agent
• Staff Role

21
Analysis Diagrams

• MESSAGE Notations
• Level 0 Analysis
• Level 1 Analysis

22
MESSAGE Notations

• Symbol Summary
• Usage of Relations

23
Symbol Summary
24
Usage of Relations

• Implication
  Assignment
• DataFlow

25
Level 0 Analysis

• Organization View
• Goal/Task View

26
Organization View

• Level 0 Organization Diagram (Structural
  Relationships)
• Level 0 Organization Diagram (Acquaintance
  Relationships)

27
Goal/Task View

• Level 0 Goal/Task Implication Diagram
  Level 0 Workflow D. of Receiving Information
• Level 0 Workflow D. of Delivering Information
  Level 0 Workflow D. of Replacing Information

28
Goal/Task View

• Level 0 Workflow D. of Receiving Question
  Level 0 workflow D. of Understanding Question
• Designating the Staff
• Level 0
  Workflow D. of Forwarding
• 
  Student to Staff

29
Goal/Task View

• Level 0
  Workflow D. of Finding Answer
• Level 0 Workflow D. of
  Offering Answer to End User

30
Level 1 Analysis

• Organization View
• Agent/Role View
• Interaction View
• Domain View

31
Organization View

• Level 1 Organization D.
  (Acquaintance Relationship)

32
Agent/Role View

• Level 1
  Delegation Structure Diagram

33
Interaction view

• Interaction D. (Student broker)
  Interaction D. (Student Staff)
• Interaction D. (Administrator Staff)

34
Domain View

• Domain Information Diagram

35
Design Process

• Agent-Platform Oriented Approach
• Architecture Driven Approach
• Design Activities

36
Design Activities

• Refining the Analysis Entities
• Selecting the Agent Architecture
• Structuring the Results according to the
  Organization View
• Organization View in the Analysis
• Organization View in the Design
• Fulfilling the Architecture
• Identifying components for External Behaviour
• Fulfilling the Domain Layer
• Fulfilling the Decision and Management Layer

37
Refining the Analysis Entities

• Interaction between Student Broker
  Student Staff
• 
  Staff Administrator

38
Identifying components for External Behaviour

• Generate Interfaces
• Generate Proxies
• Generate servers
• Generate Session Entities

39
Identifying components for External Behaviour

• Resource Layer for Interface agent Broker
  agent
• Staff Agent

40
Fulfilling the Domain Layer

• Domain Entity Mangers
• Tasks

41
Implementation

• Java Aglet API (J-AAPI) and Aglet Object Model
• Aglet Event Model
• Aglet API
• Implementation Description of Registrar Office

42
Java Aglet API (J-AAPI) and Aglet Object Model

• Main Elements
• Aglet
• Context
• Proxy
• Message
• Itinerary
• Identifier
• Supported Behaviour
• Creation
• Cloning
• Dispatch
• Retract
• Activate/Deactivate
• Dispose
• Messaging

43
Aglet Event Model

• Clone Listener
• Mobility Listener
• Persistence Listener

44
Aglet API

• Aglet class
• AgletProxy Interface
• Aglet Context
• Message Class

45
Implementation Description of Registrar Office

• Interface Agent
• Broker Agent
• Staff Agent

46
Refrences

• 1. Danny B. Lang, Mitsuru Oshima, Mobile Agents
  with Java The Aglet API, 1998
  www.moe-lange.com/danny/wwwj.pdf.
• 2. Danny B. Lang, Mitsuru Oshima, Gunter Karjoth,
  Kazuya Kosaka, Aglets Programming Mobile Agents
  in Java,1996 http//www.trl.ibm.co.jp/aglets.
• 3. WWW.sourceforge.net/projects/aglets.
• 4. Giovanni Caire, Francisco Leal, Paulo Chainho,
  Richards Evans, Francisco Garijo, Jorge Gomez,
  Juan Pavon, Paul Kearney, Jamie Stark, Philippe
  Massonet, Agent Oriented Analysis using
  MESSAGE/UML, 2000
• http//www.cediti.be/download/software_agent.
  pdf .
• 5. MESSAGE Methodology for Engineering Systems
  of Software
• Agents Methodology for agent-Oriented
  Software agents,
• www.eurescom.de/projects/p907, September
  2001.
• 6. MESSAGE Methodology for Engineering Systems
  of Software Agents, Recommendations on
  Supporting Case Tools, www.eurescom.de/projec
  ts/p907.
• 7. Natalya F. Noy, Deborah L. McGuinness,
  Ontology Development 101 A Guide to Crating Your
  First Ontology
• http//protege.stanford.edu/publications/ontol
  ogy_development/ontology101-noy-mcguinness.html.
• 8. Joseph P. Bigus, Jennifer P. Bigus,
  Constructing Intelligent Agents using Java, 2nd
  Edition, JOHN WIELY SONS INC, 2001.