運用具有問題導向的初始機制與簡群演算法求解動態武器－目標分配問題

Simplified Swarm Optimization for The Dynamic Weapon-Target Assignment Problem

10.29893/NCUTMAN.201811.0012

賴智明(Chyh-Ming Lai)；吳從華(Tsung-Hua Wu)

重層嚇阻（multi-domain deterrence） ； 武器－目標分配（weapon-target assignment, WTA） ； 動態武器目標分配（dynamic WTA） ； 簡群演算法（simplified swarm optimization） ； 基因演算法（genetic algorithm） ； multilayer defense ； multiple threats ； weapon-target assignment problem ； simplified swarm optimization ； genetic algorithm

管理學術研討會

第十六屆（2018 / 11 / 01）

57 - 67

繁體中文

去年，國防部提出新軍事戰略：「防衛固守，重層嚇阻」，作為國土防衛作戰時的軍事指導，其中「重層嚇阻」可視為達成「防衛固守」的主要手段。當面對敵軍多重來襲威脅時，我軍將聯合陸、海、空軍，運用多種武器系統編成火網，藉多層火力逐步削弱敵軍進攻能量，以達防衛固守之目的。有效的火力分配是部署多層攔截火網的成功關鍵，涉及軍事作業研究中典型的武器－目標分配問題。本研究採用目標為基的動態武器目標分配模型，應用新興的簡群演算法並輔以一個確定性的初始機制，探討、分析國土防衛作戰時多層攔截火網「武器－目標分配」的最佳化問題。藉由18組隨機想定，評估演算法的效能，並與優異的基因演算法、粒群演算法及群基式模擬退火演算法等進行比較。結果顯示，簡群演算法在求解品質與效率上均有優異的表現。

When Taiwan's homeland defense begins, the offense projects a variety of weapons (as targets to defender) at valuable assets of Taiwan to facilitate the achievement of taking over the island. According to the military strategic guideline of Taiwan, all available weapons, which constitute a multilayer defense system, are assigned to counteract these targets to eliminate the multiple threats. Effective weapon assignment is the key to success for this multilayer defense system and involves the weapon-target assignment problem (WTA) which is a classical constrained combinatorial optimization problem arising the field of military operations research. In this work, simplified swarm optimization (SSO) with a novel initialization scheme is adopted to deal with the dynamic WTA. To empirically evaluate the performance of the proposed method, experiments are conducted using eighteen randomly generated scenarios, and the corresponding results are compared with genetic algorithm, particle swarm optimization and simulated annealing. The comparison results indicate that SSO is better than its competitors in both solution quality and efficiency.