基于多影响因素的加权任务分配策略  

Weighted Task Assignment Strategy Based on Multiple Influence Factors

作　　者：尚福华[1] 孙胜男[1] 陈效果 

机构地区：[1]东北石油大学计算机信息技术学院,大庆163318 [2]东方地球物理勘探有限公司,涿州072750

出　　处：《计算机与数字工程》2014年第7期1160-1164,共5页Computer & Digital Engineering

基　　金：国家自然科学基金(编号:61170132);黑龙江省教育厅科学技术研究项目(编号:12521055)资助

摘　　要：为了解决工作流中人工型任务的自动优选分配问题,针对目前的任务分配算法考虑的因素比较固定,灵活性较差等缺点,在已有研究基础上,详细分析任务分配的影响因素,提出一种基于多影响因素的加权任务分配策略。该策略首先根据参与者的状态有效性和负载状况划分资源候选者集合,然后从资源候选者集合中按照综合匹配度进行执行者集合划分,将任务分配给评估分数最高的执行者集合,同时采用计时机制,完成任务的再次分配。同时,给出任务分配策略的形式化定义,以及分配算法的详细步骤。通过实验验证了该方法的合理性与适用性。In order to solve the problem of automatic assignment of manually tasks in workflow ,for the present task as-signment algorithms consider quite fixed factors ,have less of flexibility ,a multi-factors weighted task allocation strategy is proposed based on existing research .First ,resource candidates set is divided based on state availability and load conditions of actors ,then executor candidates set is divided with a comprehensive matching degree in the resource candidates set ;tasks are assigned to the executor candidates set which has the highest evaluate score .Meanwhile ,a tming mechanism is used to realize the redistribution of task .The formal definition of task assignment strategy and detail steps of assignment algorithm is given at the same time .Finally ,the rationality and adaptation of the method are verified through an experiment .

关 键 词：工作流 任务分配 影响因素 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]