Developing scenario-oriented function modules in new product development projects based on big data-driven approach
提出一种利用客户在线评论数据、结合关联规则挖掘和网络建模的面向场景功能模块开发方法，通过改进的GN聚类算法划分功能网络，并以智能家居系统为例验证，为新产品开发提供管理启示。
In new product development (NPD), one of the fundamental challenges is developing function modules based on usage scenarios. To enhance customer experience and satisfaction, this paper proposes a scenario-oriented approach to function modular development that leverages customer online reviews as a data source. First, to measure the scenario-oriented dependency strength among functions extracted from online reviews, we develop a method that integrates association rule mining with network modelling (i.e. a ‘scenario-function' network), incorporating scenario importance and scenario-function dependencies. Next, to evaluate the importance of scenarios, we introduce a model based on network weighted degree centrality and propose a method for quantifying the dependency strength between scenarios and functions. Finally, to partition the function network, we propose a modified GN clustering algorithm that incorporates edge weights and network information entropy. A smart home system is used as an example to illustrate the proposed big data-driven model. The results yield several managerial insights, such as methods for assessing the importance of scenarios and identifying the functions that should be incorporated into a module (or cluster).