摘要
针对物流配送过程中客户需求较大、单辆车难以满足且客户对驶入车型有限制的问题,构建了多车型需求可拆分车辆调度模型。提出交互烟花算法(interactive fireworks algorithm,In-FWA)优化求解模型,在烟花算法(fireworks algorithm,FWA)的基础上作出如下改进:1)在FWA常规爆炸中引入一种新的扇形爆炸机制,加强了烟花间的信息交流;2)使用螺旋变异方式替换了原有的高斯变异,加入了变异烟花与当前最优烟花间的信息交流;3)采取种群间精英群体吸取较差群体中较优维度的方式,进一步加强了算法种群间的信息交互性。最后,通过实验进行验证并与增强型烟花算法(enhanced fireworks algorithm,EFWA)、粒子群优化(particle swarm optimization,PSO)算法、免疫粒子群优化(immune particle swarm optimization,IPSO)算法、蚁群算法(ant colony algorithm,ACA)进行对比分析。结果表明In-FWA具有更高的收敛速度与局部搜索精度,明确了In-FWA求解该问题的有效性与优越性,并提出基于本算法的车辆调度方案。
In view of the problem that the customer demand is large and single vehicle is difficult to meet in the process of logistics distribution,and that the customer has restrictions on the driving model,a multi-type vehicle split delivery vehicle scheduling model was constructed.An interactive fireworks algorithm(In-FWA)was proposed to solve the model.Based on the fireworks algorithm(FWA),the In-FWA makes the following improvements:1)Introducing a new fan-shaped explosion mechanism into the conventional explosion of FWA,which strengthens the information exchange between fireworks;2)Replacing the original Gaussian variation with the spiral variation method,and adding the information exchange between the mutant fireworks and the current optimal fireworks;3)Adopting the way that the elite groups among the populations absorb the better dimensions in the poorer groups to enhance the information interactivity between the populations in the algorithm.Finally,experiments were carried out to verify and compare with enhanced fireworks algorithm(EFWA),particle swarm optimization(PSO)algorithm,immune particle swarm optimization(IPSO)algorithm,ant colony algorithm(ACA).The results show that the In-FWA has higher convergence speed and local search accuracy,which makes clear the effectiveness and superiority of the In-FWA to solve the problem,and a vehicle scheduling scheme was proposed based on this algorithm case.
作者
王素欣
刘浩伯
卢福强
温恒
WANG Suxin;LIU Haobo;LU Fuqiang;WEN Heng(College of Information Science and Engineering,Northeastern University at Qinhuangdao,Qinhuangdao,Hebei 066004,China)
出处
《中国科技论文》
CAS
北大核心
2019年第11期1192-1197,1222,共7页
China Sciencepaper
基金
国家自然科学基金资助项目(71401027)
中央高校基本科研业务费专项资金资助项目(N172304016)
河北省自然科学基金资助项目(G2016501086)
关键词
车辆调度
多车型
需求可拆分
烟花算法
交互烟花算法
vehicle scheduling
multi-type vehicle
split delivery
fireworks algorithm(FWA)
interactive fireworks algorithm(In-FWA)
