Topic 9: MAS - Conclusion

1
Topic 9MAS - Conclusion

• pitfalls of MAS SE
• MAS as essentially SW architecture

2
Pitfalls of Agent Development

• MAS is a philosophy
• Developing MAS is in essence software
  engineering
• how to engineer MAS is work in progress
• abstractions
• Pitfall categories
• political
• management
• conceptual
• analysis and design
• micro (agent) level
• macro (society) level
• implementation

Software Engineering with Agents Pitfalls and
Pratfalls M. Wooldridge, N. Jennings (http//cites
eer.ist.psu.edu/522355.html )
3
MAS as essentially SW architecture

• UP Main Activities
• requirements / analysis
• research the problem domain
• design towards implementable artefacts
• implementation

software architecture
4
Architectural Analysis

• common steps in architectural analysis
• identification of architectural drivers ?
  the science part
• architectural solutions ? the art part
• the science part
• identification of non-functional requirements
• in the context of functional requirements
• (e.g. security for a sales application)
• the art part
• recording architectural alternatives, decisions
  and motivations
• priorities
• describes HOW the software solution will be built
• to meet the high-priority requirements

5
Essence of MAS

• MAS is in essence
• a solution strategy
• a basis for a software architecture
• for systems with high-priority to
  (non-functional) requirements such as
• adaptability
• reliability
• scalability
• in a
• distributed world
• very dynamic world
• ? flexible and adaptive to change
• NOT FOR FREE !
• owning a hammer does not make one an
  architect
• SWA an important step in architectural analysis

6
MAS Issues and Challenges

• How do we decompose problems into behaviour for
  individual agents ?
• How do we ensure agents act coherently in making
  decision or taking action ?
• do local actions have harmful global effects
• avoiding unstable system behaviour
• How to enable agents to communicate and interact
  ?
• communication languages and protocols
• interoperation of heterogeneous agents
• finding useful existing agents in open
  environments
• How does agent decide what to do ?
• action selection mechanisms
• How do we build agents ?
• actions, plans, and knowledge
• coordination actions

7
Questions ?