Product launch success prediction model: a case study of CienaLab

The developed artifact is an interactive decision-support system designed to evaluate the performance of new product launches in the streetwear segment. It integrates machine learning techniques with visual and material attributes of apparel items, classifying each product into performance categories (best performer, hyped, or normal) based on historical sales patterns. The system processes the brand’s dataset, performs automatic encoding of categorical variables, identifies the most suitable predictive model, and generates performance-class predictions for new products. It also incorporates an interactive graphical interface that allows users to input product attributes, estimate success probabilities, and simulate entire collections. Through visual insights, intelligent color-classification rules (Wang + LAB mapping), automated consistency checks, and scenario-testing modules, the artifact enables users to analyze the impact of design attributes, compare alternatives, and support product development decisions. Designed as a practical managerial tool, it enhances decision accuracy and strengthens data-driven processes in fashion product launch planning.