PurposeThe purpose of this paper is to develop a dress design knowledge base (DDKB), which is expected to be further applied to a personalized dress recommendation system.Design/methodology/approachDress design knowledge can be expressed as the relationship between designer's fashion perceptions of different dress elements. In order to extract dress design knowledge, a dress shape ontology (DSO) is firstly developed, which can be further used to form a dress element matrix (DEM). A perceptual descriptive space of the dress (DPDS) is developed for the description of the designer's fashion perception of dress. Through a standard sensory evaluation procedure performed by experienced experts (designers), the expected relationship can be obtained. This relationship is then mathematically simulated by fuzzy logic tools for the expected DDKB.FindingsIn this paper, a DDKB has been developed. The established knowledge base has been validated, and it can be further applied to dress recommendation system for a specific consumer.Originality/valueThis study introduces the concept of knowledge base to the area of dress individualized design. The knowledge-based design process based on sensory evaluation and fuzzy logic can efficiently solve the individualization of dress design in traditional design processes, which can provide a novel way to dress design individualization.

Globally, customers are getting increasingly demanding in terms of personalization of products and are asking for shorter product development periods with more predictable product performance, especially in fashion industry. Current market pressures drive firms to adapt new design process in product development (PD) processes. Nevertheless, choosing the effective PD process is a challenging, complex decision. There is a critical need to develop a performance measurements system (PMS) for choosing appropriate product development (PD) processes in garment design to support product mangers to effectively respond to market. This paper presents a knowledge-based open performance measurement system (KBO-PMS) in big data environment, in order to support complex industrial decision-making for new product development. Its dynamic and flexible structure enables the whole system to be more adapted to knowledge sharing of product managers and processing of various time-varying data. The proposed KBO-PMS is composed of an interactive structure, capable of both integrating new KPIs from the open resource and tracking the evolution of the KBO-PMS components with time. The proposed KBO-PMS has been validated by realizing the performance evaluation of product development (PD) in fashion industry. It can be regarded as an application of open-resource based dynamic group decision-making in fashion big data environment. INDEX TERMS Dynamic key performance indicators, dynamic group decision-making, knowledge-based system, fashion big data, open system.