The Influence of Weather on Fashion Retail: Developing a Sustainable and Interpretable Forecasting Model

Ledesma Gonzalez, Cesar Antonio (2025) The Influence of Weather on Fashion Retail: Developing a Sustainable and Interpretable Forecasting Model. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (4MB) | Preview
Preview	PDF (Configuration Manual) Download (3MB) | Preview

Abstract

This study explores how weather affects sales across U.S. fashion retailer, but more than that, it is an attempt to build a sustainable, transparent forecasting architecture that balances predictive performance and ethical considerations. Led by a focus on actionable forecasting and environmental protection, the analysis integrates and refines sales transactions with weather records, applying a Linear Regression, XGBoost and Facebook Prophet, and then adding interpretability via SHAP. Based on the findings, the most streamlined approach, Linear Regression achieves the highest coefficient of determination (R2 = 0.61), underscoring the comparative significance of carefully curated input features and intuitive modelling over more intricate, complex systems. Weather data showed only modest predictive lifts; in contrast, attributes reflecting product hierarchy and calendar structure proved more impactful. A distinctive contribution is the embedding of CO2 emissions forecasting into model performance metrics, filling a conspicuous gap in the evolving literature on sustainable AI models. The prototype functions simultaneously as a forecasting engine for the fashion domain and as a modest step toward AI systems that are accountable, transparent, and ecologically minded. The work is premised on the conviction that data-driven analysis can advance corporate objectives while respecting environmental constraints, honouring the minimal ethical duty of practitioners to reduce the adverse impacts of the technologies they design and deploy.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Del Rosal, Victor UNSPECIFIED
Uncontrolled Keywords:	Sales Forecasting; Prophet; Explainable AI; SHAP; Fashion Retail; Responsible AI
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HF Commerce > Marketing > Consumer Behaviour H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Fashion Industry
Divisions:	School of Computing > Master of Science in Artificial Intelligence for Business
Depositing User:	Ciara O'Brien
Date Deposited:	24 Jun 2026 11:28
Last Modified:	24 Jun 2026 11:28
URI:	https://norma.ncirl.ie/id/eprint/9401

Actions (login required)

View Item