考虑新技能学习机制的软件项目调度人工蜂群算法  

Artificial bee colony algorithm for software project scheduling considering new skills learning mechanism

作　　者：申晓宁[1,2,3] 姚铖滨 徐继勇 宋丽妍 王玉芳[1,2,3] SHEN Xiao-ning;YAO Cheng-bin;XU Ji-yong;SONG Li-yan;WANG Yu-fang(School of Automation,Nanjing University of Information Science and Technology,Nanjing 210044,China;Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology,Nanjing University of Information Science and Technology,Nanjing 210044,China;Jiangsu Key Laboratory of Big Data Analysis Technology,Nanjing University of Information Science and Technology,Nanjing 210044,China;Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation,Southern University of Science and Technology,Shenzhen 518055,China)

机构地区：[1]南京信息工程大学自动化学院,南京210044 [2]南京信息工程大学、江苏省大气环境与装备技术协同创新中心,南京210044 [3]南京信息工程大学、江苏省大数据分析技术重点实验室,南京210044 [4]南方科技大学、广东省类脑智能计算重点实验室,广东深圳518055

出　　处：《控制与决策》2023年第3期790-796,共7页Control and Decision

基　　金：广东省重点实验室项目(2020B121201001);国家自然科学基金项目(61502239,62002148);江苏省自然科学基金项目(BK20150924)。

摘　　要：考虑新技能的学习机制,建立软件项目调度问题的数学模型.该模型融入员工对新技能的学习、新技能熟练度的增长、投入度的自适应变化以及已有技能熟练度变化等实际因素,通过寻找最佳员工任务分配方案,最小化软件项目的工期和成本.为求解该模型,提出一种引入问题启发信息的离散人工蜂群算法.将多元学习策略应用于引领蜂阶段,在保证种群多样性的同时,加强算法全局搜索能力.在跟随蜂阶段采用一种基于启发信息的变异机制,保留最优个体中契合度较高的员工信息,并根据不同个体目标值的优劣采用相异的变异方式,针对性地进行搜索,以增强算法的局部寻优能力.实验结果表明,与已有算法相比,所提算法在不同规模的软件项目调度问题中均能够搜索到更优的分配方案.Considering the learning mechanism of new skills, a mathematical model of the software project scheduling problem is established. The model integrates some practical factors such as learning of new skills, increase of the new and existing skill proficiencies, and adaptive changes of dedications. Both duration and cost of the software project are minimized by finding the best assignment of employees to tasks. To solve the model, a discrete artificial bee colony algorithm incorporating heuristic information is proposed. A multi-learning strategy is applied to the employed bees phase to enhance the global search ability of the algorithm while maintaining the population diversity. In the onlooker bees phase, a mutation mechanism based on heuristic information is adopted, where information of the employees with higher fit in the optimal individual is retained, and distinct mutation operators are employed on different individuals based on their objective values to improve the local search ability. Experimental results show that compared with the existing methods, the proposed algorithm can find a better allocation in software project scheduling problems with increasing scales.

关 键 词：软件项目调度 新技能学习机制 熟练度 人工蜂群算法 多元学习 启发信息 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]