The idle time which is part of the order fulfillment time is decided by the number of items in the zone; therefore the item assignment method affects the picking efficiency. Whereas previous studies only focus on the ...The idle time which is part of the order fulfillment time is decided by the number of items in the zone; therefore the item assignment method affects the picking efficiency. Whereas previous studies only focus on the balance of number of kinds of items between different zones but not the number of items and the idle time in each zone. In this paper, an idle factor is proposed to measure the idle time exactly. The idle factor is proven to obey the same vary trend with the idle time, so the object of this problem can be simplified from minimizing idle time to minimizing idle factor. Based on this, the model of item assignment problem in synchronized zone automated order picking system is built. The model is a form of relaxation of parallel machine scheduling problem which had been proven to be NP-complete. To solve the model, a taboo search algorithm is proposed. The main idea of the algorithm is minimizing the greatest idle factor of zones with the 2-exchange algorithm. Finally, the simulation which applies the data collected from a tobacco distribution center is conducted to evaluate the performance of the algorithm. The result verifies the model and shows the algorithm can do a steady work to reduce idle time and the idle time can be reduced by 45.63% on average. This research proposed an approach to measure the idle time in synchronized zone automated order picking system. The approach can improve the picking efficiency significantly and can be seen as theoretical basis when optimizing the synchronized automated order picking systems.展开更多
针对入侵检测中数据特征维度高的问题,提出了改进粒子群联合禁忌搜索(IPSO-TS)的特征选择算法。采用遗传算子对粒子群算法进行了改进,得到了特征选择初始最优解;对该解进行禁忌搜索(TS)得到了特征子集的全局优化解。基于KDD CUP 99数据...针对入侵检测中数据特征维度高的问题,提出了改进粒子群联合禁忌搜索(IPSO-TS)的特征选择算法。采用遗传算子对粒子群算法进行了改进,得到了特征选择初始最优解;对该解进行禁忌搜索(TS)得到了特征子集的全局优化解。基于KDD CUP 99数据集的实验结果表明,相较遗传算子整合粒子群算法(CMPSO)、粒子群算法(PSO)和粒子群联合禁忌算法,IPSO-TS减少了至少29.2%的特征,缩短了至少15%的平均检测时间,提高了至少2.96%的平均分类准确率。展开更多
基金Supported by Independent Innovation Foundation of Shandong University of China(Grant No.2013GN007)
文摘The idle time which is part of the order fulfillment time is decided by the number of items in the zone; therefore the item assignment method affects the picking efficiency. Whereas previous studies only focus on the balance of number of kinds of items between different zones but not the number of items and the idle time in each zone. In this paper, an idle factor is proposed to measure the idle time exactly. The idle factor is proven to obey the same vary trend with the idle time, so the object of this problem can be simplified from minimizing idle time to minimizing idle factor. Based on this, the model of item assignment problem in synchronized zone automated order picking system is built. The model is a form of relaxation of parallel machine scheduling problem which had been proven to be NP-complete. To solve the model, a taboo search algorithm is proposed. The main idea of the algorithm is minimizing the greatest idle factor of zones with the 2-exchange algorithm. Finally, the simulation which applies the data collected from a tobacco distribution center is conducted to evaluate the performance of the algorithm. The result verifies the model and shows the algorithm can do a steady work to reduce idle time and the idle time can be reduced by 45.63% on average. This research proposed an approach to measure the idle time in synchronized zone automated order picking system. The approach can improve the picking efficiency significantly and can be seen as theoretical basis when optimizing the synchronized automated order picking systems.
文摘针对入侵检测中数据特征维度高的问题,提出了改进粒子群联合禁忌搜索(IPSO-TS)的特征选择算法。采用遗传算子对粒子群算法进行了改进,得到了特征选择初始最优解;对该解进行禁忌搜索(TS)得到了特征子集的全局优化解。基于KDD CUP 99数据集的实验结果表明,相较遗传算子整合粒子群算法(CMPSO)、粒子群算法(PSO)和粒子群联合禁忌算法,IPSO-TS减少了至少29.2%的特征,缩短了至少15%的平均检测时间,提高了至少2.96%的平均分类准确率。
