To decarbonise road transport, charging infrastructure must be accessible, reliable, and economically viable. OMV built AI into its EV business from day one to make that happen, from site selection to grid optimization.
Scaling electric mobility is not as simple as installing more chargers. Each new charging point must operate within tight grid constraints, uncertain demand patterns, and a clear commercial framework. Electricity capacity must be secured in advance, capital must be deployed carefully, and infrastructure must perform reliably over time.
At OMV, the e‑mobility business is growing within these real‑world constraints. As the charging network expands, data‑driven analytics – and increasingly, artificial intelligence – are becoming important tools to support better decisions, manage complexity, and prepare the business for scale.
Road transport accounts for around 15% of global energy‑related emissions, and the International Energy Agency (IEA) identifies electric vehicles as a key technology to reduce those emissions. Adoption is accelerating rapidly, with electric vehicles now accounting for more than one in five new car sales worldwide. This growth puts pressure on charging infrastructure to expand in a way that remains economically viable and system‑compatible.
E‑mobility is a strategic growth area within OMV’s Strategy 2030. More than 1,200 charging points are already in operation across Austria, Romania, Slovakia, Hungary, and Bulgaria, with further expansion planned by 2030. Delivering this growth requires disciplined investment and a clear understanding of where digital tools can add real value.
Using data and analytics to support better decisions
From the early phases of network expansion, OMV has relied on data and advanced analytics to support investment and operational decisions. OMV is advancing AI through targeted applications and pilots building a clear understanding of where machine-based learning can deliver measurable value.
Location decisions: grounding expansion in data
Choosing the right locations is critical for any charging network. Successful sites combine technical feasibility, customer demand, and a viable business case. Poorly chosen locations risk low utilization and inefficient use of capital.
OMV started with its existing network of filling stations and applied data‑driven analysis to assess which sites could realistically host EV chargers.
Jelena Brborović
Department Manager for Advanced Analytics & AI at OMV
Analytics support the early screening of sites by highlighting constraints such as limited grid capacity, insufficient space, or challenging utility connections. Beyond these basic criteria, teams analyze market characteristics, customer behavior, and competitive landscapes such as proximity to destinations where drivers typically spend time and the availability of nearby charging options.
Rather than replacing expert judgment, these insights help teams focus their attention on the most promising locations and reduce uncertainty in investment decisions.
Preparing for growing power demand
As charging networks grow, electricity demand becomes a central operational challenge. In Austria, operators must secure grid capacity in advance to ensure system stability. Underestimating demand can lead to penalties, while overestimating can tie up capacity and capital unnecessarily.
Today, OMV relies on a combination of operational data, experience, and analytical tools to forecast demand. At the same time, AI‑based models are being piloted to explore whether more granular forecasting – incorporating factors such as traffic patterns or usage trends – can improve accuracy over time.
These pilots are an important step toward understanding how machine learning could support electricity procurement and grid bookings as the network scales. Improvements in forecasting could help free up capacity and improve flexibility for future expansion.
From pilots to operational support
Beyond planning, OMV is also exploring how advanced analytics and AI could support day‑to‑day operations. One example is the use of pattern recognition to help identify irregular charging behavior, supporting fraud prevention and customer trust.
Another area under evaluation is energy optimization. As OMV pilots solar generation and battery storage at selected charging sites, the energy system becomes more complex. Analytical models are being tested to assess how different sources of electricity could be balanced efficiently while maintaining reliability.
Predictive maintenance is another aspect. While most maintenance today remains reactive, data‑driven approaches are being explored to identify recurring patterns that could signal potential issues earlier.
“The goal is to better understand where data can help us intervene earlier and improve availability and reliability, as demand grows” says Brborović.
Scaling pragmatically
OMV’s e‑mobility journey illustrates a pragmatic approach to digitalization. AI is not positioned as a silver bullet, but as a set of tools that, when applied selectively, can support better decisions, more efficient operations, and disciplined growth.
As OMV continues to expand its low‑carbon portfolio, many technologies will need to transition from pilot projects to scalable, profitable businesses. Advanced analytics and AI will play an increasingly important role in that transition – but only where they demonstrate clear value.
By combining operational experience with data‑driven insight, OMV is building e‑mobility infrastructure that can scale responsibly supporting the decarbonization of road transport, while maintaining reliability, performance, and economic discipline.