Adoption of Machine Learning in the Development of a Manufacturing Quality Control Framework

在制造质量控制框架的开发中采用机器学习

• 批准号: 525092-2018
• 负责人: Blum, Susan
• 金额: $ 2.91万
• 依托单位: Saskatchewan Polytechnic
• 依托单位国家: 加拿大
• 项目类别: Applied Research and Development Grants - Level 2
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=645657
• 关键词: Adoption; Machine; Learning; Development; Manufacturing

The Mercedes-Benz Fuel Cell (MBFC) division of Mercedes-Benz Canada Inc. is responsible for the research, development and initial production of a hydrogen fuel cell driven motor for the Mercedes-Benz vehicle lineup. MBFC has developed a system to identify and remove substandard parts from their fuel cells early in the manufacturing process. Early removal of substandard parts decreases the time and resources spent by MBFC on continued manufacturing of parts that will subsequently be removed from the process. MBFC plans to improve their already rigorous quality control measures building on the work started with an Engage project. MBFC is continuing their partnership with Saskatchewan Polytechnic's Digital Integration Centre of Excellence (DICE) to improve manufacturing data analysis through the use of unsupervised and semi-supervised data analytics. Improvements to the existing early identification system will identify and flag substandard parts for removal early in the manufacturing process, which will provide cost savings and increase the final quality of hydrogen fuel cells.

梅赛德斯-奔驰加拿大公司的梅赛德斯-奔驰燃料电池（MBFC）部门。负责梅赛德斯-奔驰汽车系列氢燃料电池驱动电机的研究、开发和初始生产。 MBFC已经开发了一个系统，可以在制造过程的早期识别和去除燃料电池中的不合格部件。尽早去除不合格零件减少了MBFC在继续制造随后将从工艺中去除的零件上花费的时间和资源。 MBFC计划在Engage项目开始的工作基础上改进其已经严格的质量控制措施。 MBFC正在继续与萨斯喀彻温省理工学院的数字集成卓越中心（DICE）合作，通过使用无监督和半监督数据分析来改善制造数据分析。对现有早期识别系统的改进将在制造过程的早期识别和标记不合格的零件，以节省成本并提高氢燃料电池的最终质量。