Learning Objectives
- Explain how Cemex’s customer-service strategy evolved from digitizing the order-to-fulfillment journey with Cemex Go to embedding AI into core business processes.
- Recognize why AI at scale requires more than successful models, including reliable data foundations, reusable platforms, business ownership, change management, and ongoing model operations.
- Examine how Cemex organized digital enablement, IT, global data science, HR, and business process leaders to balance experimentation with enterprise-grade deployment.
- Assess how Cemex prioritized data science opportunities through its Speed to Value framework, distinguishing globally scalable “integrated” solutions from locally valuable “coordinated” solutions.
- Analyze how Cemex industrialized AI by recontextualizing models across markets, explaining model behavior through Magic Tools, managing model drift, and adopting MLOps practices.
Abstract
Headquartered near Monterrey, Mexico, in 2023 Cemex, S.A.B. de C.V. (Cemex) was a $15.6 billion global construction materials company that employed 43,000 people and managed a global supply chain with distribution networks in more than fifty countries. In 2014, Fernando González became CEO and introduced a strategic imperative to deliver superior customer service, resulting in Cemex Go, the first end-to-end digital platform in the construction materials industry. As Cemex established its digital capabilities, the company increasingly built and relied on AI to fuel automated, intelligent business processes. By 2023, the company had incorporated into Cemex Go a collection of interrelated artificial intelligence (AI) models that fueled core components of the order-to-fulfillment process. This working paper describes Cemex’s journey to managing AI at Scale, defined as the organizational ability to sustain large numbers of interdependent, production AI models over time. The working paper features the key activities that facilitated the company’s AI at Scale progress.
Appropriate for the Following Course(s)
artificial intelligence, digital transformation, data management and analytics, organizational design, data monetization, and IT investment and portfolio management
THERE IS NO TEACHING NOTE FOR THIS CASE STUDY.