Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Decision Maker Model as Multiobjective Optimization Problem for Genetic Algorithm
Transcript of Decision Maker Model as Multiobjective Optimization Problem for Genetic Algorithm
supervisor Dr.sc.ing. Asoc. Prof. Janis Eiduks AICT 2010, April 22-23 The role of Decision Maker in optimization Decision Maker (DM) have significant impact on solving optimization problems by using interactive optimization methods
In optimization procedure DM provide preference information for optimization method
DM decide to continue or stop a solution search process
Optimization procedure Find initial solution Decision
Maker A problem of Multi-objective optimization Multiobjective
optimization method Data and charts Start Stop i=i+1 i=1 DM's preference information
others Pareto optimal solutions (compromise solutions) All Pareto optimal solutions are nondominated solutions Difficulties of testing interactive optimization methods The goal is to determine performance of new and classic interactive methods Classic testing of methods Modern testing of methods large DM count, usualy >65
objective and subjective of metrics
large workload of experiment
use value functions for measure performance
easy to plan experiments
metrics has mathematical type
it is not necessary use the real DM Model of Decision Maker Defined as multiobjective optimization problem (two criteria)
Based on evolutionary algorithms (stohastic type)
Is the part of testing ground (tool) Testing Ground 1 2 3 preference information F1(X)- Euclidean distance to the goal solution
F2(X)- Correctness of preference information Multiobjective Optimization Genetic Algorithm (MOGA) next slide 4 5 6 Results of some experiment Problem -> Process of solution search -> Monitoring MOGA-> 7 Conclusions It is not necessary to use real Decision Makers for method testing
Testing ground (tool) provide all necessary functionality of testing methods in manual or in automatic (by using model of Decision Maker) mode
In future we need to compare this suggested model with real Decision Maker's behaviour in solving optimization problems
For some optimization problems the decision time is more than 5 min
By using stochastic type of Model of Decision Maker we can collect experiment data and analyse its by statistical methods