关键词: 需求预测, 服装市场, 预测模型, 知识图谱, 研究热点

Abstract: Through a visual analysis of research in the field of clothing demand forecasting at home and abroad, this study aims to understand relevant research hotspots and development trends, offering reference for future research in this area domestically. Based on relevant literature from the core database of Web of Science (WoS) and China National Knowledge Infrastructure (CNKI) on clothing demand forecasting, research literature from 2004 to 2024 was collected and analyzed. By using the bibliometric analysis software CiteSpace, statistical analysis was conducted across multiple dimensions such as the evolutionary process and keyword clustering, and knowledge maps were drawn to analyze the research overview, hotspots, and development trends in the field of clothing demand forecasting. The results show that the relevant research primarily focuses on the following four main areas. The first area is product development and fashion forecasting, which includes predictions of consumers' clothing preferences and fashion trends, as well as research on how fashion trend forecasting can effectively guide the clothing design process. The second area is supply chain management and demand forecasting. The research focuses on exploring how to optimize supply chain management processes and mitigate risks such as overproduction and stockouts through demand forecasting, and categorizes demand forecasting methods based on the popularity level of apparel products. The third area is brand clothing and sales forecasting, where relevant research primarily predicts the sales performance of brand clothing based on market data and consumer behavior. The fourth area is consumer psychology and behavior forecasting, where research focuses on analyzing how to incorporate consumers' personal preferences, psychological factors, purchasing behaviors, etc., into forecasting models to improve prediction accuracy. Finally, the future research directions in the field of clothing demand forecasting are summarized. Future research can unfold in multiple directions. The first is the development of intelligent forecasting systems, where researchers can enhance the intelligence level of forecasting systems by introducing artificial intelligence and machine learning technologies. Second, refining forecasting models is also an important direction for future research, requiring models to improve their accuracy and adaptability. Meanwhile, researchers should consider integrating multiple data sources such as social media, e-commerce platforms, and consumer behavior to ensure data diversity, so as to enhance the accuracy of forecasting models. Lastly, with the increasingly diverse preferences of consumers, personalized demand forecasting that accurately predicts the specific needs and preferences of different customer groups will play an increasingly important role in optimizing sales forecasts and driving the success of apparel businesses.

Key words: demand forecasting, clothing market, prediction model, knowledge map, research hotspot