Product Line Optimization for Fast Fashion Retailers

This research project considers the operations of a 'fast-fashion' retailer such as Zara or H&M. With the speed at which the fashion industry moves, such retailers have developed and invested in merchandize procurement strategies that permit lead times as short as two weeks. A level of flexibility is required that allows retailers to adjust the assortment of products offered on sale at their stores quickly enough to adapt to popular fashion trends. In particular, such retailers use weekly sales data to renew their estimates specific items’ popularity, and based on such revised estimates, weed out unpopular items, or else re-stock demonstrably popular ones on a week-by-week basis. In sharp contrast, traditional retailers such as Marks and Spencer face lead times on the order of several months. As such these retailers need to predict popular fashions months in advance and are allowed virtually no changes to their product assortments over the course of a sales season, which is typically several months in length. Understandably, this approach is not nearly as successful at identifying high selling fashions and also results in substantial unsold inventories at the end of a sales season. In view of the great deal of a-priori uncertainty in the popularity of a new fashion and the speed at which fashion trends evolve, the fast-fashion operations model is highly desirable and emerging as the de-facto operations model for large fashion retailers.

Candidates for this project should have obtained a first university degree with a substantial quantitative component, ideally a good degree with a mathematical or statistical major/minor. The research will involve computer programming, most likely in MatLab. Training and technical support will be given as necessary.

For further particulars please contact the supervisor, Joern Meissner.