266 fuzzy front-end (FFE) studies in the new product development (NPD) sector were examined. The studies were selected using a bibliometrics method, and chronologically and statistically examined with ten criteria divided into two dimensions. The first dimension is associated with overall attributes of the FFE, consisting of six criteria: the study taxonomy, model type, NPD speed, NPD attributes, model characteristic, and model structure. The second dimension is relevant to the FFE performance structure related to process parameters, comprised of four criteria: the FFE task, activity, performance method, and toolkit. In terms of those two dimensions, the paper looks at previous FFE studies to gain an understanding of features of each FFE study along with related knowledge and theories, as well as identification of evolution trends of FFE studies. Based on the identification, an FFE model development strategy for each criterion is formulated, and this paper proposes possible options for executing those strategies which exert influence on the form of the cluster network. The intention is for the database to be utilised as an overview of all existing FFE studies and allow specific FFE studies to be selected to examine FFE approaches.This paper provides FFE model development guidance on how to deal with the overall attributes and outcomes of the FFE which affect the entirety of the innovation process, and how to manage the performance structure related to process parameters.

审查了新产品开发（NPD）领域的266个模糊前端（FFE）研究。使用文献计量学方法选择研究，并按十个标准（分为两个维度）按时间顺序和统计方式进行检查。第一个维度与FFE的总体属性相关，包括六个标准：研究分类法，模型类型，NPD速度，NPD属性，模型特征和模型结构。第二个维度与与过程参数相关的FFE绩效结构有关，它由四个标准组成：FFE任务，活动，绩效方法和工具包。在这两个方面，本文着眼于以前的FFE研究，以了解每个FFE研究的特征以及相关的知识和理论，并确定FFE研究的发展趋势。在识别的基础上，针对每个准则制定了FFE模型开发策略，并为执行那些对集群网络的形式产生影响的策略提出了可能的选择。目的是将数据库用作所有现有FFE研究的概述，并允许选择特定的FFE研究来检查FFE方法。本文为如何处理FFE的整体属性和结果提供了FFE模型开发指南。这会影响整个创新过程，以及如何管理与过程参数相关的绩效结构。并且本文提出了执行那些对集群网络形式产生影响的策略的可能选择。目的是将数据库用作所有现有FFE研究的概述，并允许选择特定的FFE研究来检查FFE方法。本文为如何处理FFE的整体属性和结果提供了FFE模型开发指南。这会影响整个创新过程，以及如何管理与过程参数相关的绩效结构。并且本文提出了执行那些对集群网络形式产生影响的策略的可能选择。目的是将数据库用作所有现有FFE研究的概述，并允许选择特定的FFE研究来检查FFE方法。本文为如何处理FFE的整体属性和结果提供了FFE模型开发指南。这会影响整个创新过程，以及如何管理与过程参数相关的绩效结构。