Towards a solver-aware systems architecting framework: leveraging experts, specialists and the crowd to design innovative complex systems

Towards a solver-aware systems architecting framework: leveraging experts, specialists and the crowd to design innovative complex systems
复制标题

迈向求解器感知系统架构框架:利用专家、专家和大众来设计创新的复杂系统

DOI:
10.1017/dsj.2022.7
复制
发表时间:
2022
期刊:
影响因子:
2.4
通讯作者:
Lifshitz-Assaf, Hila
Lifshitz-Assaf, Hila
中科院分区:
--
文献类型:
--
作者:
Szajnfarber, Zoe;Topcu, Taylan G.;Lifshitz-Assaf, Hila

文献摘要

参考文献

相似文献

本文提出了解决方案感知的系统架构框架,利用专家,群众和专家的综合优势,设计创新的复杂系统。虽然系统架构理论已经广泛地探讨了可供选择的架构形式和性能之间的关系下运行的不确定性,有限的注意力已经支付的差异,由于谁产生的解决方案。最近兴起的替代解决方法,从零工到众包,再到新颖的合同结构,都强调了在复杂系统创新的背景下,需要更深入地考虑架构和解决方案能力之间的联系。我们通过一个抽象的解决问题的模拟,代表不同的专业知识,从事通过合同结构,匹配他们的解决类型的替代分解和求解器原型,调查这些相互作用。我们发现,首选架构的变化取决于求解器的组合被分配。此外,最好的混合分解求解器组合同时提高了性能和成本,同时减少了对专家的依赖。为了操作这个新的求解器感知框架,我们诱导两个分解分配对的算法,并在模拟中展示其值的规模。我们还应用这两个逻辑推理的机器人机械手设计问题的一个例子,以证明其在现实的复杂系统设置的相关性。
This article proposes the solver-aware system architecting framework for leveraging the combined strengths of experts, crowds and specialists to design innovative complex systems. Although system architecting theory has extensively explored the relationship between alternative architecture forms and performance under operational uncertainty, limited attention has been paid to differences due to who generates the solutions. The recent rise in alternative solving methods, from gig workers to crowdsourcing to novel contracting structures emphasises the need for deeper consideration of the link between architecting and solver-capability in the context of complex system innovation. We investigate these interactions through an abstract problem-solving simulation, representing alternative decompositions and solver archetypes of varying expertise, engaged through contractual structures that match their solving type. We find that the preferred architecture changes depending on which combinations of solvers are assigned. In addition, the best hybrid decomposition-solver combinations simultaneously improve performance and cost, while reducing expert reliance. To operationalise this new solver-aware framework, we induce two heuristics for decomposition-assignment pairs and demonstrate the scale of their value in the simulation. We also apply these two heuristics to reason about an example of a robotic manipulator design problem to demonstrate their relevance in realistic complex system settings.
DOI: 10.1115/1.4040548
发表时间: 2018-07
影响因子: 3.3
作者:
Murtuza N. Shergadwala;Ilias Bilionis;Karthik N. Kannan;Jitesh H. Panchal
通讯作者: Murtuza N. Shergadwala;Ilias Bilionis;Karthik N. Kannan;Jitesh H. Panchal
DOI: --
发表时间: 2016
期刊:
影响因子: --
作者:
Jessica Daecher
通讯作者: Jessica Daecher
医学中的角色和创新
DOI: --
发表时间: 1960
影响因子: 4.4
作者:
Joseph Ben
通讯作者: Joseph Ben
在工程设计中利用群体——迈向整体框架
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Jitesh H. Panchal
通讯作者: Jitesh H. Panchal
知识距离、认知搜索过程和创造力
DOI: --
发表时间: 2016
期刊: Psychology Science
影响因子: --
作者:
O. A. Acar;J. van den Ende
通讯作者: J. van den Ende