Explainable AI based Interventions for Pre-season Decision Making in Fashion Retail

AI can reduce fashion waste through better demand forecasting: The paper addresses a key sustainability challenge in fashion—overproduction of inventory—by using AI to better understand consumer shopping behavior and predict which new products will succeed, thereby reducing the environmental impact of unsold goods.

Explainability is essential for creative industries to adopt AI: The authors emphasize that in fashion design and merchandising, stakeholders are more likely to trust and use AI tools when they can understand why the system makes specific recommendations, rather than treating it as a "black box"—a finding particularly important for domains that rely on human creativity and intuition.

AI tools can bridge competing stakeholder priorities: The paper proposes an "interventional analysis" approach that allows different team members—from data-driven buyers and financial planners to intuition-based designers—to collaborate in product decisions by making AI recommendations transparent enough for all parties to understand and challenge them.

Explainability enables human oversight in high-stakes decisions: By building interpretability into the forecasting tool, the system allows stakeholders to participate in decision-making rather than passively accepting AI predictions, which is crucial for responsible AI deployment in business contexts where decisions affect both company performance and resource consumption.