New product development to enhance companies’ competitiveness and reputation is one of the leading activities in manufacturing. At present, achieving successful product design has become more difficult, even for companies with extensive capabilities in the market, because of disorganisation in the fuzzy front end (FFE) of the innovation process. Tremendous amounts of information, such as data on customers, manufacturing capability, and market trend, are considered in the FFE phase to avoid common flaws in product design. Because of the high degree of uncertainties in the FFE, multidimensional and high-volume data are added from time to time at the beginning of the formal product development process. To address the above concerns, deploying big data analytics to establish industrial intelligence is an active but still under-researched area. In this paper, an intelligent product design framework is proposed to incorporate fuzzy association rule mining (FARM) and a genetic algorithm (GA) into a recursive association-rule-based fuzzy inference system to bridge the gap between customer attributes and design parameters. Considering the current incidence of epidemics, such as the COVID-19 pandemic, communication of information in the FFE stage may be hindered. Through this study, a recursive learning scheme is established, therefore, to strengthen market performance, design performance, and sustainability on product design. It is found that the industrial big data analytics in the FFE process achieve greater flexibility and self-improvement mechanism on the evolution of product design.

开发新产品以增强公司的竞争力和声誉是制造业的主要活动之一。目前，由于创新过程的模糊前端（FFE）混乱，即使对于市场上具有广泛能力的公司而言，成功实现产品设计也变得更加困难。在FFE阶段要考虑大量信息，例如有关客户，制造能力和市场趋势的数据，以避免产品设计中的常见缺陷。由于FFE中存在高度不确定性，因此在正式产品开发过程开始时会不时添加多维和大容量数据。为了解决上述问题，部署大数据分析以建立工业智能是一个活跃但仍处于研究不足的领域。本文提出了一种智能产品设计框架，将模糊关联规则挖掘（FARM）和遗传算法（GA）结合到基于递归关联规则的模糊推理系统中，以弥合客户属性和设计参数之间的差距。考虑到当前流行病的发生（例如COVID-19大流行），在FFE阶段的信息交流可能会受到阻碍。通过这项研究，建立了一个递归学习方案，以增强产品设计的市场绩效，设计绩效和可持续性。研究发现，FFE流程中的工业大数据分析在产品设计的演进中实现了更大的灵活性和自我完善机制。提出了一种智能产品设计框架，将模糊关联规则挖掘（FARM）和遗传算法（GA）结合到基于递归关联规则的模糊推理系统中，以弥合客户属性和设计参数之间的差距。考虑到当前流行病的发生，例如COVID-19大流行，在FFE阶段的信息交流可能会受到阻碍。通过这项研究，建立了一个递归学习方案，以增强产品设计的市场绩效，设计绩效和可持续性。研究发现，FFE流程中的工业大数据分析在产品设计的演进中实现了更大的灵活性和自我完善机制。提出了一种智能产品设计框架，将模糊关联规则挖掘（FARM）和遗传算法（GA）结合到基于递归关联规则的模糊推理系统中，以弥合客户属性和设计参数之间的差距。考虑到当前流行病的发生（例如COVID-19大流行），在FFE阶段的信息交流可能会受到阻碍。通过这项研究，建立了一个递归学习方案，以增强产品设计的市场绩效，设计绩效和可持续性。研究发现，FFE流程中的工业大数据分析在产品设计的演进中实现了更大的灵活性和自我完善机制。