Hidden order: how adaptation builds complexity
Hidden order: how adaptation builds complexity
Introduction to artificial life
Introduction to artificial life
Hierarchical federations: an architecture for information hiding
Proceedings of the fifteenth workshop on Parallel and distributed simulation
Trends in Cooperative Distributed Problem Solving
IEEE Transactions on Knowledge and Data Engineering
Efficient and Scalable Pareto Optimization by Evolutionary Local Selection Algorithms
Evolutionary Computation
Adaption in distributed systems: an evolutionary approach
Proceedings of the 8th annual conference on Genetic and evolutionary computation
| Hi-index | 0.00 |
We propose a decentralized evolutionary approach for studying autonomous heterogeneous agents interacting in a supply chain. Such logistics networks can be seen as complex networks that need to adapt their internal structure (e.g. transport routes, interactions) as reaction to environmental changes, e.g. the market demand, supplier unavailability or route changes. We model such distributed supply chains as a decentralized multi-agent system in order to draw an analogy to real world scenarios. This paper describes a decentralized evolutionary optimization approach that differs in two ways from traditional EA. First the fitness calculation is replaced by an economic model. Second the entire agent population constructs only one solution. The connections in supply-chains can be seen as a complex network of coexisting but simple interdependent agent strategies producing together the necessary transportation network. We describe how our decentralized approach can be used to solve inherently distributed problems where no central optimization algorithm exist. The simulation results show the applicability of the approach to transport network optimization.