BY LARS PERSSON
Although artificial intelligence has existed as a concept since the 1950s, it wasn’t until the 2010s that the technology began to be seriously integrated into modern, commercial supply chain software. In recent years, there has been rapid development that has refined the ability to plan, optimize, and make decisions. Two particularly exciting areas are conversational AI and the development of AI agents.
The first modern AI-based supply chain systems developed in the 2010s enabled us, among other things, to make accurate demand forecasts and optimize routes in real time. Today, AI systems can also anticipate problems, simulate alternatives, provide recommendations, and independently plan and make decisions. Tasks that previously often required specialized expertise and manual analysis processes can now be performed by AI agents that initiate complex calculations and run simulations using powerful optimization algorithms.
Interactive AI Agents
At PTV Logistics, we recently launched PTV Mira, which is a great example of advancements in AI. PTV Mira is based on more than four decades of experience in developing routing and optimization technologies and performs real-world simulations that generate optimized, accurate results. In this solution, we combine AI with various types of optimization techniques and use a conversational interface. This new approach is similar to how many of us use tools like ChatGPT, Copilot, and Claude.
Data-driven, optimized responses
Asking questions in PTV Mira triggers sophisticated processing behind the scenes. Examples of questions include:
”What happens if our shipping volumes increase by 18 percent on Wednesday?”
”Show me the least efficient routes in this delivery plan?”
”Should we open our new warehouse in Helsingborg or in Malmö?”
You explain a problem just as you would when talking to a colleague. No technical expertise is required. The system then processes your question, performs the necessary calculations, and immediately delivers accurate, data-driven results. As your collaboration with your new assistant and expert deepens, the tasks can become more advanced, progressing from providing explanations to making recommendations and then to taking action.
Route Optimization and Planning
PTV Mira has access to the core functions of our software platform, including route planning and optimization, mapping, modeling for electric vehicles, consideration of road restrictions, and scenario creation and analysis. Based on this access, PTV Mira can orchestrate routes, generate alternative scenarios, and compare different outcomes in real time—that is, it can support both short-term operational issues and more strategic decisions such as planning depot locations or conducting sustainability analyses.
Professional, fast, and accurate
In other words, PTV Mira is much more powerful than a chatbot that generates text-based responses based on its historical data. You and your colleagues can focus on asking the right questions and verifying the results and recommendations provided. In short, focus on what people do better than machines. PTV Mira does not replace planners, analysts, or decision-makers; rather, it supports them in their work.
But what about the risks?
When people talk about AI, questions about risks often come up. And even though the risks of inaccurate analyses and recommendations are negligible, I believe that we humans need to have a good understanding of how AI agents work, what their limitations are, and make sure to verify the results. If we do that, these new digital coworkers will open up a world of possibilities. PTV Mira is ready to be your assistant or consultant!
—
Lars Persson is Vice President of Sales for the Nordics and the UK at PTV Logistics
