Adriana Costache, Service Management Lead, Michelin CBS Bucharest: We are committed to building AI systems that are sustainable, with a focus on minimizing resource consumption and environmental impact

Digital transformation has long been a key priority for Michelin Corporate & Business Services (CBS). It affects every service line, bringing both challenges and successes—some shared across the organization, others unique to specific teams. To better understand what this transformation has meant during such a pivotal time, we talked to Adriana Costache, Service Management Lead, Michelin CBS Bucharest  to share her and the compay’s perspectives:

Which high-impact business processes have you prioritized for AI and automation in 2025, and how are you measuring success?

At CBS Bucharest, we don’t just automate – we transform. Our journey with AI and automation in 2025 has been about more than technology. It’s been about freeing people from repetitive work, making smarter decisions, and building a more sustainable future.

Let’s start where many digital journeys begin, with repetitive, manual tasks that drain time and energy. For us, one such example is the Accounts Payable process for one of our product units – each month, the team handled manually a significant number of invoices, a process that was slow and error-prone. In response, we built an AI-powered solution combining document understanding and robotic process automation (RPA). Soon, invoices will be automatically read, validated, and processed, which means we will have faster payment cycles, less errors, and significantly better experience for both employees and suppliers.

Another compelling transformation came from our employee tire benefit program. Previously, it required multiple emails, manual validations, and long waiting times. Now, our chatbot Alex, integrated into Microsoft Teams, handles the entire flow. Employees simply chat with Alex, who verifies their eligibility, confirms tire stock and even places the order. Behind the scenes, an RPA bot takes care of the heavy lifting. The result? A 24/7, fully compliant, and user-friendly service that saves time and improves satisfaction.

These stories are part of a broader, deliberate strategy which we’ve deployed until now: more than 30+ live bots, 3 chatbots, real-time business intelligence through Power BI and Databricks, 49+ Citizen Developers, trained to build their own automations and solve local challenges.

In 2025, we are expanding into Generative AI, with three pilot projects currently in the pipeline. These focus on automated invoice validation using cross-referenced data from multiple systems for logistic domain, finance knowledge agents that support intragroup flows, and AI copilots for Order-to-Cash processes, designed to streamline documentation and decision support. These projects aim to further accelerate accuracy, compliance, and responsiveness in core operations, amplifying the work of our teams with smarter tools, not just faster ones.

We track the success of our AI and automation initiatives using a blend of quantitative and qualitative KPIs, focused on delivering measurable business value while advancing our sustainability goals. Key metrics focus on the number of hours saved through automation, the delivered value, the level of adoption and engagement, as well as process performance.

How are you ensuring responsible and ethical AI deployment, including bias mitigation, data privacy, and regulatory compliance?

At Michelin, the principles guiding the AI systems, along the entire stages (design, development, deployment, procurement, and operation) are fully aligned with our core values as outlined in our Code of Ethics.

The Michelin Code sets out, though not exhaustively, the foundational principles that govern our approach to AI, including ownership and accountability, accuracy, robustness, and safety, ongoing risk management, continuous monitoring and remediation, security and compliance with data protection laws, AI regulations, respect for fundamental rights and individual freedoms.

In parallel, we are committed to building AI systems that are sustainable, with a focus on minimizing resource consumption and environmental impact.

When it comes to working with AI suppliers, we carefully select partners who share our values and standards. We ensure they commit, through strong contractual agreements, to key areas such as data protection, confidentiality, and intellectual property rights.

Have you implemented an AI governance framework, and how does it support transparency, auditability, and growth?

We’ve established a comprehensive AI governance framework that ensures responsible innovation while enabling scalability.

This framework includes having strategic committees which oversee prioritization and alignment with business goals. From ideation to deployment, each AI initiative follows a structured funnel, ensuring feasibility, business value, and ethical compliance are assessed early. We further track AI maturity, accessibility, and value realization through centralized portfolios and KPIs such as lead time, readiness, and actual vs. potential value. We built AI communities to democratize AI usage while maintaining oversight and consistency.

This governance model not only ensures compliance with internal and external standards but also fosters a culture of trust, collaboration, and continuous learning across regions and functions.

What role does AI-powered analytics play in driving innovation, and how are you balancing investment in AI with long-term sustainability goals?

Data is everywhere but without direction, it’s just noise. With tools like Power BI and Databricks, we’ve built dashboards that help teams across Europe make smarter, faster decisions. Whether it’s tracking industrial indicators, managing inventory discrepancies, or analyzing workforce trends, our analytics are designed to empower, not overwhelm.

One of our proudest innovations is Normie, an AI chatbot that helps employees navigate over finance documents. It uses advanced language models to deliver the right answer, instantly. It’s fast, accurate, and always available. Aurora, our in-house AI assistant and productivity partner, helps employees automate small tasks, find information faster, and make better decisions with less effort, by democratizing AI use.

We also use analytics to predict and prevent issues in supply chains and operations, benchmark performance across factories and functions, and support ESG goals with sustainability metrics.

And we do it all with a clear focus on responsibility, making sure that our projects are selected based on ROI and ESG impact, we use low-code platforms to reduce our IT footprint, and each deployment embeds ethical AI principles.

Because for us, AI isn’t just about automation, it’s about amplifying human potential.