Feasibility study of large-scale mass customization 3D printing framework system with a case study on Nanjing Happy Valley East Gate

At present, the development and implementation of digital transformation are the keys to promoting high-quality industry development. The new digital fabrication method of robotic 3D printing is a research area being studied by many to tackle the issue of the declining productivity of traditional construction methods. Although many studies have been done, most of the current 3D printing projects are facing limitations in terms of scale. In order to bridge the gap, this article proposed a mass customization 3D printing framework system for large-scale projects. This article discusses how mass customization is made possible through the joint operation of the FUROBOT software and 3D printing hardware. By taking the east gate of Nanjing Happy Valley Plaza as a case study, the article demonstrates and studies the feasibility of the large-scale mass customization 3D printing framework system.     

Mass customization of medical devices and implants: state of the art and future directions

The medical field is one in which the need for customization can be clear cut, as providing tailored devices and implants for unique physiologies can provide for a better overall treatment than the use of 'off the shelf' devices and implants. Customization in the production of medical products can be roughly divided into consideration of medical devices, and of implantable parts or systems. The present paper outlines the current state of the art in both of these areas, presents details of projects that are ongoing at the University of Leeds and outlines future research directions.     

提高定制产品物料需求预测的准确性

提高定制产品物料需求预测准确性是提高企业竞争能力的必由之路。通过某公司提高定制产品物料需求预测准确性的具体项目实践，介绍了关于提高物料需求预测准确性的具体措施。通过建立一种能为各部门所共同接受的产品语言来加强各部门的信息交流，并减少从客户的实际需求到物料预测之间的中间环节来提高整个物料预测的准确性，并以实例验证其有效性。     

面向大规模定制的虚拟设计研究

综述了大规模定制的范围经济性和面向大规模定制的基本设计方法，并提出了一个虚拟设计框架，该框架改变了传统的设计模式，利用先进的设计方法学和仿真支撑技术为产品开发提供了最大柔性和敏捷性。同时，它提供了一个多学科合作的设计环境，用户直接参与进设计循环中。最后，结合新型剑杆织机的开发，展示了一个能够支持新产品快速成功开发的虚拟设计环境。     

Nasal Epithelial Cell-Based Models for Individualized Study in Cystic Fibrosis

The emergence of highly effective CFTR modulator therapy has led to significant improvements in health care for most patients with cystic fibrosis (CF). For some, however, these therapies remain inaccessible due to the rarity of their individual CFTR variants, or due to a lack of biologic activity of the available therapies for certain variants. One proposed method of addressing this gap is the use of primary human cell-based models, which allow preclinical therapeutic testing and physiologic assessment of relevant tissue at the individual level. Nasal cells represent one such tissue source and have emerged as a powerful model for individual disease study. The ex vivo culture of nasal cells has evolved over time, and modern nasal cell models are beginning to be utilized to predict patient outcomes. This review will discuss both historical and current state-of-the art use of nasal cells for study in CF, with a particular focus on the use of such models to inform personalized patient care.     

面向网络化定制的产品族复合建模技术

在分析了网络化定制关键技术的基础上，针对性地对信息映射方法和产品族建进行了研究。提出了多层结构信息映射模型，以及综合类元件和参数化特征建模的双层复合产品族模型。结合本方法和以往网络化制造领域中所取得的成果，实现了较完整的基于Web的个性化定制过程。     

基于信息资源聚合的农民专业合作社信息系统

针对农民专业合作社信息获取与信息发布难、成本高的问题,提出基于XM L Schema的农民专业合作社信息聚合元数据表示规范,设计数据传输流程和信息定制与智能推送机制,建设两层部署体系的农民专业合作社信息系统。系统信息服务中心与农业搜索引擎（AgSoso）相结合,为远端用户系统提供最新的信息资源;系统客户端实现合作社生产经营信息的本地化管理,并向信息服务中心同步最新的农产品等信息。应用结果表明,该方案投入成本、维护成本低,资源利用效率高,具有较高的实用应用价值,为农民专业合作社提供了一套基于多来源数据聚合服务的技术解决方案。     

基于数据分布的聚类联邦学习

联邦学习(federated learning)可以解决分布式机器学习中基于隐私保护的数据碎片化和数据隔离问题。在联邦学习系统中,各参与者节点合作训练模型,利用本地数据训练局部模型,并将训练好的局部模型上传到服务器节点进行聚合。在真实的应用环境中,各节点之间的数据分布往往具有很大差异,导致联邦学习模型精确度较低。为了解决非独立同分布数据对模型精确度的影响,利用不同节点之间数据分布的相似性,提出了一个聚类联邦学习框架。在Synthetic、CIFAR-10和FEMNIST标准数据集上进行了广泛实验。与其他联邦学习方法相比,基于数据分布的聚类联邦学习对模型的准确率有较大提升,且所需的计算量也更少。     

基于slope one算法改进评分矩阵填充的协同过滤算法研究

为解决协同过滤算法中的数据稀疏性问题,提出了一种改进的协同过滤算法。该算法使用slope one算法计算出来的评分预测值来填充评分矩阵中的未评分项目,然后在填充后的用户-项目评分矩阵上通过基于用户的协同过滤方法给出推荐。利用slope one算法计算出来的评分预测值作为回填值,既能降低评分矩阵的稀疏性,也保证了回填值的多样性,从而减少均值、中值等单一填充值造成的推荐误差。在MovieLens-1M数据集上对本文改进算法和协同过滤算法及均值中心化处理的算法作五折交叉实验,结果表明,基于评分预测值填充数据后的协同过滤算法有效的缓解了数据稀疏性问题,并且有更好的推荐效果。     

一种基于移动用户位置的网络服务推荐方法

伴随着无线通信技术和智能移动终端的快速发展,基于位置的服务(location-based services,简称LBS)以其移动性、实用性、随时性和个性化的特点,在军事、交通、物流等诸多领域得到了广泛的应用,成为最具发展潜力的移动增值业务之一。在一个基于位置的网络服务推荐框架的基础上,给出了一种基于位置的移动用户偏好相似度计算方法,同时证明了其满足近邻相似测度的一般性质;然后,提出一种符合社会学概念的信任值计算方法。把它们应用于基于移动用户位置的网络服务推荐过程中,从而形成了一种基于移动用户位置的网络服务推荐方法。该方法有效地提高了网络服务推荐的准确性和可靠性,同时缓解了推荐过程中可能存在的数据稀疏性以及冷启动问题。最后,通过公开的MIT数据集验证了该推荐方法的准确度和可行性。