GBU College Logo
Sign in with GBU Microsoft

Explainable neural algorithms for corporate sustainability forecasting: A layered predictive model anchored in executive awareness, green finance, and digital innovation

Innovation and Green Development | 2026

Paper Details

Authors: Ibrahim Y.; Moubarak H.; Badawy H.

DOI: 10.1016/j.igd.2026.100335

Journal: Innovation and Green Development

Year: 2026

Publisher: Elsevier B.V.

Document Type: Article

Open Access: All Open Access; Gold Open Access

Cited by: 2

Abstract

This study investigates how artificial intelligence (AI) capability drives sustainable performance through the mediating role of Digital Green Innovation (DGI). Grounded in the Resource-Based and Natural Resource-Based Views, survey data from 321 organizations are analyzed using a multi-method approach that integrates partial least squares structural equation modeling (PLS-SEM), machine learning (ML), and explainable AI (XAI). The PLS-SEM results reveal a full mediation effect AI Capability enhances sustainable performance exclusively through DGI highlighting that technological resources must be embedded within innovation processes to generate environmental and social value. To ensure convergent validation and methodological robustness, predictive ML models (random forest, support vector regression, multilayer perceptron, and one-dimensional convolutional neural networks) are applied alongside XAI techniques (SHAP and LIME). These complementary analyses independently converge on the same key drivers DGI and top management environmental awareness providing strong empirical triangulation and interpretive transparency. Theoretically, the study advances the understanding of AI-enabled sustainability by demonstrating that AI resources yield value only when channeled through green innovation capabilities. Methodologically, it contributes by showcasing a convergent SEM–ML–XAI framework that enhances both explanatory and predictive validity. Practically, organizations should strengthen digital innovation systems and employ XAI tools to dynamically monitor and refine sustainability performance drivers. © 2026 The Authors

Keywords

AI capability; Digital green innovation; Machine learning forecasting; Model interpretability; SEM; Sustainable performance