A Social Media Powered Preseason Production Planning System for Fast-fashion New Releases

Description

Fast fashion, an emerging retail strategy of adapting frequent merchandise assortments, has been widely deployed in the fashion industry. However, due to dynamic changes in fashion trends and limited historical sales records for fashion new releases, it remains a challenging task to provide preseason production decisions for fast fashion retailers. In this project, we propose to utilize the marketing power and information cues of social media content to develop a preseason production planning system. The project first explores the predictive power of user-generated fashion content on social media platforms to build a preseason demand predictor. A data-driven production optimization algorithm is then proposed to provide preseason production decisions to minimize the new-season demand-supply mismatch. This data-driven production planning system that integrates data science and operations research can benefit the fashion retailers for preseason production planning before enough sales records are available.