Efficient multi-objective evolutionary algorithms for solving the multi-stage weapon target assignment problem: A comparison study

The weapon target assignment (WTA) problem is a fundamental problem arising in defense-related applications of operations research. The multi-stage weapon target assignment (MWTA) problem is the basis of the dynamic weapon target assignment (DWTA) problem which commonly exists in practice. The MWTA...

Full description

Saved in:
Bibliographic Details
Published in:2017 IEEE Congress on Evolutionary Computation (CEC) pp. 435 - 442
Main Authors: Juan Li, Jie Chen, Bin Xin, Lu Chen
Format: Conference Proceeding
Language:English
Published: IEEE 01.06.2017
Subjects:
combinatorial optimization
Decision making
decomposition
Discrete wavelet transforms
Evolutionary computation
multi-objective constrained optimization problem
multi-objective optimization
multi-stage weapon target assignment (MWTA)
Optimization
Upper bound
Weapons
ε-constraint
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The weapon target assignment (WTA) problem is a fundamental problem arising in defense-related applications of operations research. The multi-stage weapon target assignment (MWTA) problem is the basis of the dynamic weapon target assignment (DWTA) problem which commonly exists in practice. The MWTA problem considered in this paper is formulated as a multi-objective constrained combinatorial optimization problem with two competing objectives. Apart from maximizing the damage to hostile targets, this paper follows the principle of minimizing the ammunition consumption. Decomposition and Pareto dominance both are efficient and prevailing strategies for solving multi-objective optimization problems. Three competitive multi-objective optimizers: DMOEA-εC, NSGA-II, and MOEA/D-AWA are adopted to solve multi-objective MWTA problems efficiently. Then comparison studies among DMOEA-εC, NSGA-II, and MOEA/D-AWA on solving three different-scale MWTA instances are done. Three common used performance metrics are used to evaluate the performance of each algorithm. Numerical results demonstrate that NSGA-II performs best on small-scale and medium-scale instances compared with DMOEA-εC and MOEA/D-AWA, while DMOEA-εC shows advantages over the other two algorithms on solving the large-scale instance.
DOI:10.1109/CEC.2017.7969344