TY - GEN
T1 - TIEA
T2 - 7th International Symposium on Computational Intelligence and Industrial Applications, ISCIIA 2016
AU - Chen, Lu
AU - Xin, Bin
AU - Chen, Jie
N1 - Publisher Copyright:
© 2016, Fuji Technology Press. All rights reserved.
PY - 2016
Y1 - 2016
N2 - This paper proposes a tradeoff based interactive multiobjective evolutionary algorithm (TIEA). It can support the decision maker (DM) to guide the search towards the DM's most preferred solution by combing evolutionary algorithms with the framework of gradient projection. At each iteration, an evolutionary algorithm is used to generate an approximate Pareto optimal solution. Then, the DM is asked to provide preferences in terms of indifference tradeoffs, the projection of which onto the tangent hyperplane of the Pareto front provides a tradeoff direction. To calculate the projection, an approach to approximate the normal vector of the tangent hyperplane of the Pareto front is developed. Results on nine numerical examples by using TIEA show that TIEA is able to solve multiobjective optimization problems with more than three objectives. Moreover, it can find the Pareto optimal solution that the DM prefers most.
AB - This paper proposes a tradeoff based interactive multiobjective evolutionary algorithm (TIEA). It can support the decision maker (DM) to guide the search towards the DM's most preferred solution by combing evolutionary algorithms with the framework of gradient projection. At each iteration, an evolutionary algorithm is used to generate an approximate Pareto optimal solution. Then, the DM is asked to provide preferences in terms of indifference tradeoffs, the projection of which onto the tangent hyperplane of the Pareto front provides a tradeoff direction. To calculate the projection, an approach to approximate the normal vector of the tangent hyperplane of the Pareto front is developed. Results on nine numerical examples by using TIEA show that TIEA is able to solve multiobjective optimization problems with more than three objectives. Moreover, it can find the Pareto optimal solution that the DM prefers most.
KW - Evolutionary algorithm
KW - Interactive multiobjective optimization
KW - Normal vector approximation
KW - Tradeoffs
UR - http://www.scopus.com/inward/record.url?scp=84997646335&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84997646335
T3 - ISCIIA 2016 - 7th International Symposium on Computational Intelligence and Industrial Applications
BT - ISCIIA 2016 - 7th International Symposium on Computational Intelligence and Industrial Applications
PB - Fuji Technology Press
Y2 - 3 November 2016 through 6 November 2016
ER -