精英蜂群算法及考虑利益相关者的众包定价模型

任务分配是众包流程的关键所在，也是众包价值的重要体现。从众包活动参与者即主体企业、众包平台、平台会员的视角出发，研究服务众包定价问题。在考虑会员信誉度和任务聚集度的基础上，针对含有地理因素的众包任务设计打包分配定价方案。以服务成本、任务价值、会员收益等为导向，对不同任务进行组合配置，从而设计多目标规划任务配置及定价模型，并针对该模型构建了精英蜂群算法。在精英蜂群算法中，充分利用蜜源信息并着重考虑成长性较好的蜜蜂，进而避免了局部最优问题，提高了搜索效率。通过对众包服务企业运营数据分析，获取到众包服务会员特征及任务完成相关基础信息，以此进行仿真实验。仿真结果表明通过众包任务打包定价机制，任务完成率、企业总成本、三方总收益等方面均有显著优化。综合模型及数据实验可知，众包任务在定价与发布过程中根据自身特征差异，在无差异服务中只需要考虑会员信誉度，对于具有服务差异性的任务则需要考虑打包发布。

Elitism bee colony algorithm and crowdsourcing pricing model based on stakeholders' preferences

Task assignment is the key to connect crowdsourcing process, and it is also an important embodiment of mission value. This paper studied the pricing problem of crowdsourcing model from the perspective of participants, which include enterprises, crowdsourcing platforms and members. On the basis of considering membership credibility and task aggregation degree, an allocation pricing scheme was designed for crowdsourcing tasks with geographical factor. Guided by service cost, task value, member income, a multi objective programming model was constructed by combining different tasks, and the Elitism Bee colony Algorithm was designed for the model. In this algorithm, through the use of nectar source information and considering the improvement of bees, the search efficiency is significantly improved. Then analyzing e-commerce enterprise operation data, this paper got crowdsourcing membership characteristics, task completion rate and other basic information, and carried out the simulation experiment. Simulation results show that the task packaging pricing strategy has significant advantages in task completion rate, enterprise total cost, and the three parts total revenue. Through the model and data experiments, it can be seen that crowdsourcing has its own characteristics in pricing and publishing. For nondifferentiated services, it only needs to consider the reputation of members, for differentiated services, the task assignment. needs to be packaged.