AI enables predictive automotive logistics decisions in Mexico

From anticipating delays to optimising transport routes, speakers at ALSC Mexico 2026 examined how automakers and logistics partners can turn digital tools into faster, better-informed supply chain decisions.

Automotive companies are increasingly looking to AI and digitalisation to make their supply chains more predictive, moving beyond simply tracking shipments and collecting data towards anticipating disruption and improving operational decisions. At Automotive Logistics & Supply Chain Mexico 2026, representatives from OEMs like BMW and GM – as well as logistics and technology providers – outlined how these tools are being applied to address challenges ranging from cross-border freight movements and inbound material flows to finished vehicle logistics and operational planning.

While the applications vary, a common theme emerged: technology alone does not make a supply chain more efficient. Companies need reliable, connected data and a clear understanding of the problems they want to solve before AI can deliver meaningful results. For OEMs and their logistics partners, the challenge is now to turn growing volumes of information into actionable insights, faster decisions and more resilient operations.

What’s needed to shift from reactive to predictive decision-making?

A successful digital transformation depends on having the right pillars in place to enable the move from reactive to predictive decision-making, or from tactical to strategic moves, as Bernardo Múzquiz Pezino, finished vehicle logistics senior group manager at General Motors de México, highlighted. Being able to see a step ahead allows for strategic moves, not just tactical ones, and that, he said, is important when it comes to predicting and preventing disruption.

Adrian Jennings, chief product officer at technology provider Cognosos, noted that without clear goals, the right data and a team capable of turning insights into action, technology on its own will not provide the intended results. “Adopting technology doesn’t actually improve anything,” he said. “But the technology allows you to work in new ways that are better than the old ways.”

He suggested that before any new technology is adopted, firms should identify clear problems that the technology should be introduced to address or solve. “The kinds of problems that you want to pick are the ones that require a lot of data assimilation, complex judgment calls and decision-making – ones that are very time-consuming,” he said. “Those are the ones that lend themselves best to AI.”

Pezino agreed, noting that understanding the root cause of an issue is essential to be able to use tools to make processes more predictive, digitalised or automated. “If we do not understand what the business problem is, then the utilisation of technology just becomes a beautiful dashboard or a fancy dashboard to show,” he added.

Another important factor to the successful implementation of new technologies is the ability to communicate internally between departments to maximise the impact of the technology on overall efficiency, as Nicholas Schook, vice president of automotive at international courier CNW, alluded to.

“There needs to be open dialogue and open communication between the IT team and the technology teams, as well as the operations teams and the finance teams,” Schook said. “So if we’re going to adopt AI and other technology to make our operations more efficient or our finances more efficient, we need to be providing our operations teams with the proper tools that can make them more efficient.”

Schook shared that CNW recently hired a head of AI, but it has also empowered its operators to use AI agents to improve the efficiency of processes like routing and quoting.

From visibility to actionable insights

Visibility is an important step towards predictive analytics, but Jennings noted that visibility alone is no longer sufficient – businesses need to be able to interpret information, using data to identify risks and inform decisions.

He highlighted the importance of data quality and context, explaining that AI tools need reliable information about the processes and flows they are intended to support. Without this, gaps in the underlying data can undermine the quality of recommendations.

Jennings described the time between an event occurring and a decision being made as “decision latency”, arguing that AI can help reduce this by quickly turning data into insights and recommended actions. He also stressed that the technology needs to operate within appropriate parameters, with employees retaining oversight of decisions, particularly during the early stages of adoption.

BMW’s work in Mexico illustrates the importance of connecting information across supply chain partners. Raúl Gamboa, head of logistics, production control and production systems at BMW Group Plant San Luis Potosí, outlined how the automaker has developed digital tools to improve coordination between suppliers, logistics providers, customs brokers and other stakeholders. Its X-Border platform, for example, connects partners involved in export shipments, providing information on component requirements, supplier readiness and collection arrangements before trucks arrive.

BMW has also consolidated several tools into a single dashboard, bringing together information such as track-and-trace and shipment status. These platforms provide the connected information needed to improve visibility and support more predictive decision-making.

AI applications in inbound and vehicle logistics

Speakers at ALSC Mexico 2026 outlined several practical applications of AI across inbound material movements and finished vehicle logistics, particularly where better forecasting could help companies respond before disruption affects operations.

At BMW, Gamboa described an AI tool that recommends the most suitable and efficient transport stream for export freight, with the aim of improving truck utilisation, reducing costs and increasing efficiency. The company has also developed tools to recommend routes for expedited shipments, balancing cost and delivery requirements.

Gamboa also described the use of predictive visibility analytics to identify whether shipments are likely to arrive early or late before they enter the road network. This gives logistics teams an opportunity to respond before delays affect material availability or create wider problems for the plant.

Another potential use case for AI and digital tools, highlighted by Pezino at GM, is to improve vehicle yard planning. By mapping yard capacity and expected vehicle volumes alongside factors such as weather, quality issues, rail limitations and port conditions, teams can anticipate congestion and consider alternative transport options. “If we do the right yard mapping, for instance, we can predict that we’re going to get congested by X amount of units by whatever day,” he said.

Pezino also pointed to the use of digital images and AI to standardise vehicle quality inspections, helping improve the identification of damage and the handling of claims. The approach could make assessments more consistent while reducing the manual effort involved.

For logistics providers, AI is also being applied to day-to-day transport planning. Schook said CNW uses an AI routing agent to compare different options, suggesting – for example – the fastest route, a cheaper alternative and a slower, more economical option. This allows operators to respond quicker to customer requests while retaining the ability to assess the most appropriate option in context.

Measuring the benefits of AI applications

Although speakers at the conference described a growing range of AI applications, the evidence of measurable benefits varied between companies and use cases. Some examples involved quantified improvements, while others were focused on the potential to improve efficiency, responsiveness and decision-making.

At GM, Pezino said the company had automated the equivalent of 1,400 hours of manual work so far in 2026. He also said his team had access to more than five AI tools. Although not all of these 1,400 hours have been automated by AI specifically, the figure illustrates the potential for digital tools to reduce repetitive work and give employees more time to focus on more value-add activities.

Schook described similar benefits at CNW, where AI is being used to help operators compare routing options and respond quicker to requests from OEM and tier supplier customers. The company is also applying AI to workload forecasting, helping it anticipate busier days and hours so that operations teams can be staffed accordingly. Schook noted that AI could also be used to reduce manual data entry in billing and finance processes.

Although automating manual tasks can provide significant benefits for an organisation and AI tools can provide valuable insights, speakers at ALSC Mexico stressed that humans must remain responsible for final judgements, taking into account wider contextual factors that an AI tool might not consider. This also keeps human operators accountable for the decisions made. According to Jennings, the value of AI will depend on how effectively companies combine its recommendations with operational knowledge and human judgement.