- Job scheduling → “based on Daimler-Chrysler”
- In a factory → groups of ultra modern machines (GoM)
- Each GoM can do various groups of operations on “prefabricated piece of material”
- Typically, one GoM is not enough to complete a workflow
- Completing a product requires multiple GoMs used in a specific order
- Different requests materialize in an order that cannot be a’priori predicted
- Various requests may have different “priority levels”
- Different machines within GoMs can experience problems that will force them out of service, or require them to be (graciously) taken out of service – for limited (not necessarily known in advance) time
- In Daimler-Chrysler case it was shown that placing an agent within each GoM, and adding management agents to the ecosystem, can result in productivity improvement
- Aim of the project is to develop an agent-based “simulator” that will allow modeling and studying such ecosystem
link do schematu znajomości agentów: https://drive.google.com/file/d/1s-KRB-jUtB8HHYWPJhxboZYnUP75Y7pU/view?usp=sharing