节点文献
面向开放式设计的群体协作及决策优化研究
Research on Group Collaboration and Decision Optimization in Open Design
【作者】 杨俊;
【导师】 唐敦兵;
【作者基本信息】 南京航空航天大学 , 机械电子工程, 2020, 博士
【摘要】 开放式设计是一种基于互联网的大众参与式的产品创新设计模式。在开放式设计环境下,大量具有不同知识背景的参与者借助开放式的网络社区共同参与到产品的创新设计过程。因此,相比于传统的封闭式的产品设计模式,开放式设计极大地拓宽了产品创新的渠道。同时,开放式设计在实现技术创新、加速创意转化、节约创新成本、提高用户满意度等方面也表现出了突出优势。然而,开放式设计基础理论研究的不足严重制约了开放式设计思想在复杂产品系统研发中的应用。为此,本文从社区资源的组织协调和优化配置角度出发,对开放式设计环境下参与者大规模群体协作的组织模式以及开放式设计过程中的决策优化问题展开了研究。主要研究内容总结如下:1、对开放式设计环境下参与者大规模群体协作的组织模式和实施方法进行了研究。首先,讨论了开放式设计环境下设计平台对于外部创新资源的利用模式,指出参与者大规模群体协作是开放式设计的核心,并进一步提出了开放式设计的具体实施框架。然后,通过在FBS模型基础上引入需求和资源信息,提出了一种拓展的RFBSR模型,实现了开放式设计项目需求-功能-行为-结构-资源之间的映射。最后,基于复杂网络的概念,提出了一种支持开放式设计大规模群体协作的知识协作网络模型,将RFBSR模型构成元素及其映射关系抽象为网络中的节点和连边,并从知识和效率角度给出了知识协作网络模型的评价指标,进而通过对知识协作网络拓扑结构的优化实现开放设计过程中社区资源的组织和协调。2、对开放式设计环境下参与者大规模群体协作活动的鲁棒性问题进行了研究。首先,从网络的鲁棒性角度出发,讨论了资源节点失效对知识协作网络鲁棒性的影响,并提出利用知识协作网络的鲁棒性评价指标对开放式设计环境下参与者大规模群体协作活动的鲁棒性进行评价。然后,从知识和效率角度进一步定义了知识协作网络的鲁棒性评价指标,包括知识容量损失、知识富集度损失、经验损失和协作效率损失。最后,为了保证开放式设计环境下参与者大规模群体协作活动能够持续稳定开展,在基于知识协作网络的社区资源组织协调过程中引入了网络的鲁棒性指标,并进一步讨论了不同网络节点失效策略对于知识协作网络鲁棒性的影响,提出了对应的网络鲁棒性优化策略。3、对开放式设计环境下设计方案的快速筛选和评价问题进行了研究。首先,指出开放式设计过程中产品设计方案的筛选和评价问题是一个典型的不确定性多指标决策问题,且具有方案数量多、方案指标不确定、决策信息不完备、决策实时性要求高等特点。考虑到上述特点,提出采用随机多指标方案可接受度分析方法(SMAA)解决上述决策问题,并给出了基于SMAA的多指标决策问题求解原理。然后,从提高决策指标的可靠性、提高决策效率、降低计算复杂度等多个角度对SMAA模型进行了改进,提出了一种改进的迭代型随机多目标可接受度分析方法—ISMAA,实现了决策者决策信息缺失以及决策评价指标不确定条件下设计方案的快速筛选和评价。此外,还提出了序敏感度概念,分析了方案指标的不确定水平对于方案最终排序结果的影响,并据此对基于I-SMAA的决策实施流程进行了改进,进一步提高了决策结果的可靠性。最后,结合实际案例验证了所提方法的可行性和有效性。4、对开放式设计信息不完备条件下决策者偏好信息的提取以及设计方案的最终决策问题进行了研究。首先,考虑到在开放式设计环境下,很难完整、准确地获取决策者的偏好信息,提出将方案决策指标之间的成对比较关系视为一种粗糙集决策规则,通过随机多指标可接受度分析(SMAA)模型与考虑优势关系的变精度粗糙集(DB-VPRS)模型的结合实现上述决策规则的重要度评估。然后,基于决策规则的重要度实现了对偏好信息的优化提取以及对冗余偏好信息的剔除。最后,结合实际案例对所提方法的可行性和有效性进行了验证。5、开发了面向开放式设计的群体协作平台原型系统。首先,讨论了面向开放式设计的群体协作平台的整体系统架构和功能组成;然后,借助QT Webengine完成了面向开放式设计的群体协作平台原型系统的开发;最后,结合工程实例对上述知识协作平台原型系统的功能进行了演示和说明。
【Abstract】 Open design is an Internet-based product innovation design mode with public participation.In the open design environment,a large number of participants with different knowledge backgrounds participate in the product innovation design process with the help of the open network community.Therefore,compared with the traditional closed design mode,open design brings more innovative solutions.Meanwhile,it also has tremendous advantages in realizing technological innovation,accelerating innovation transformation,saving innovation cost and improving user satisfaction.However,the deficiency of the basic theory of open design has seriously restricted its application in complex product systems development.To this end,from the perspective of organization and coordination and optimal allocation of community resources,the organization mode of participants’ large-scale group collaboration and decision-making optimization in the process of open design are studied in this paper.The major research contents are summarized as follows.1.The organization mode and operation mechanism of participants’ large-scale group collaboration in open design environment are studied.Firstly,the utilization mode of external innovation resources in the design platform in the open design environment is discussed.It is pointed that participants’ largescale group collaboration is the core of open design.Moreover,a specific implementation framework is proposed.Secondly,by introducing requirements and resources information into the FBS model,an extended RFBSR model is proposed to realize the mapping between requirements,functions,behaviors,structures and resources in open design project.Finally,based on the concept of Complex Network,a group collaboration network model is proposed to support participants’ large-scale group collaboration in open design environment.In the RFBSR model,the elements and their mapping relations are abstracted as nodes and edges of the network.Moreover,the evaluation indexes of the group collaboration network are given from the perspective of knowledge and efficiency.In this way,the organization and coordination of community resources in open design can be realized by optimizing the topology of the group collaboration network.2.The robustness of participants’ large-scale group collaboration in open design environment is studied.Firstly,from the perspective of network robustness,the influence of resource nodes failure on the robustness of the group collaboration network is discussed,and the robustness evaluation indexes of group collaboration network are proposed to evaluate the robustness of participants’ large-scale collaboration in open design.Secondly,from the perspective of knowledge and efficiency,the robustness evaluation indexes of group collaboration network are defined,including loss of knowledge capacity,loss of knowledge enrichment,loss of experience and loss of working efficiency.Finally,in order to make the group collaboration steady and consistent in open design,the robustness evaluation indexes of group collaboration network are introduced into the organization and coordination process of the community resources.Furthermore,the influence of different node failure strategies on the robustness of group collaboration network is discussed,and the corresponding network robustness improvement strategies are proposed.3.The rapid screening and evaluation problem of design alternatives in the open design environment is studied.Firstly,it is pointed that the screening and evaluation problem of design alternatives in the open design environment is a typical multi-criteria decision-making problem with uncertainty,and it is characterized by large number of alternatives,uncertain evaluation indexes,incomplete information and high requirement for real-time decision-making.Considering the above characteristics,the stochastic multi-criteria acceptability analysis(SMAA)method is used to solve this problem,and its principle is also introduced.Then,the SMAA model is improved from the aspects of reliability,efficiency and computational complexity,and an improved iterative stochastic multi-criteria acceptability analysis method(I-SMAA)is presented.This method can help realize a rapid screening and evaluation of design alternatives with uncertain evaluation indexes and incomplete information.In addition,the concept of rank sensitivity is introduced,which is used to analyze the influence on the ranking result of alternatives caused by uncertainty level of evaluation indexes.Based on the sensitivity analysis result,the decision implementation process based on I-SMAA is further improved to increase its reliability.Finally,a case study is used to verify the performance and efficiency of the proposed method.4.The extraction of decision makers’ preference information and the decision-making of design alternatives are studied.Firstly,considering that it is difficult to obtain decision makers’ preference information completely and accurately,the pairwise comparison relationships between indexes are considered as rough decision rules.The SMAA model is combined with the DB-VPRS model to evaluated the importance of these decision rules.Then,based on the importance of decision rules,the optimal extraction of preference information and the elimination of redundant information are realized.Finally,a case study is used to evaluate the performance and efficiency of the proposed method.5.A prototype system of group collaboration for open design is developed.Firstly,the overall system architecture and functions of the group collaboration platform for open design are discussed.Then,a prototype system of the group collaboration platform for open design is developed using QT Webengine.Finally,the functions of the system are demonstrated and illustrated with a project example