Cloud-native causal AI for supply chain KPI monitoring: A GCP framework to diagnose out-of-stock events
Year: 2025
Type: article
Abstract: Effective supply chain management (SCM) is essential for ensuring operational efficiency, cost optimization, and customer satisfaction. In an increasingly dynamic business environment, the integration of cloud computing and Causal Artificial Intelligence (AI) presents new opportunities for value creation and intelligent decision-making. This study proposes a framework that leverages the capabilities of Google© Cloud Platform (GCP) for real-time Key Performance Indicator (KPI) monitoring and supply chain analytics. By incorporating Causal AI, the framework enables the identification of underlying causes of out-of-stock (OOS) situations, rather than merely observing correlations. The research presents an end-to-end architecture combining data pipelines, real-time dashboards, and causal inference models to proactively detect and address OOS risks. This integrated, data-driven approach aims to improve inventory accuracy, enhance forecasting, and ultimately strengthen supply chain resilience and responsiveness. • Cloud-native framework on GCP delivers real-time causal analytics for supply chains. • Causal AI pinpoints supplier OTIF and lead-time setup as dominant stockout drivers. • Improving OTIF by 1% lowers stockout risk by 1.21% across test data. • Requested lead-time tuning cuts stockout likelihood by 11.7%, boosting resilience. • Forecast accuracy and planner overrides showed no significant causal effect on OOS.
Authors: Tarique Ameer, Omid Fatahi Valilai
Institution: Constructor University
Language: English
Subfield: Management Information Systems
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,340
OpenAlex ID: W4416041237