Model the future, optimize the present, and turn complex decisions into measurable outcomes, using simulation, optimization, and AI.
Start a conversationOperations Research, AI, and data analysis. Three interdisciplinary fields I bridge to get the best outcome.
Help companies figure out the best way to use their resources, whether that's scheduling shifts, routing trucks, or testing "what if" scenarios to build a foundation for changes.
Build custom AI tools that actually fit real workflows, from forecasting demand to automating repetitive tasks with machine learning and large language models.
Turn messy data into clear answers. Interactive dashboards, statistical analysis, and system integrations that help teams make better decisions faster.
A mix of finished and ongoing work. Some came from client projects, others from research I've translated into tools that actually get used. What they share: digging into the details and figuring out what holds up in practice.
Built an interactive online platform to train thousands of decision-makers and practitioners for pilgrimages with over a million participants. Evidence-based training for safety, efficiency, and preparedness.
Simulated months of SKU flows across 1M+ items to optimize size limits, assignment rules, and storage strategies. Simulation-based parameters outperformed Excel planning for smoother operations from day one.
Modeled how a million fans would move through Doha across all transport modes with complex decision models. Identified critical bottlenecks and tested interventions to ensure smooth crowd flow during the tournament.
Redesigned the periodic delivery routes for over 100 libraries. The optimized solution eliminated the need for one vehicle while maintaining service levels.
Optimized schedules for mobile libraries covering 1,000+ stops across more than five depots. Balanced visit frequency, driver shifts, and vehicle capacity into one coherent plan. The planning tool behind this work became fahrbuecher.de.
Plan your routes on fahrbuecher.de →Developed an algorithm that reduced split shipments by 80% in theoretical evaluations for an e-commerce retailer. Smarter inventory placement means fewer boxes, lower shipping costs, happier customers.
Redesigned district boundaries and dispatching rules for German and Belgian police forces. Simulation-tested changes showed improved response times and coverage, with results under discussion for adoption.
Built AI tutors for university courses that guide students through problems step by step, without giving away answers. Improved understanding and engagement while maintaining academic integrity. Out of this project grew Oshu, a full platform for grounded chatbots.
Build your own chat agent →Developing intelligent routing for demand-responsive transport in rural areas. Flexible scheduling that adapts to actual passenger needs instead of empty fixed routes.
Every project is different, especially if the problem hasn't been tackled yet. What follows is the shape most of them take.
Who are you, what's stuck, what have you tried? I sit down with everyone involved, the team doing the work and the people setting direction, and map how things actually run and where the data lives. By the end we have a clear picture of what you actually need.
I structure the problem and pressure-test the assumptions. Where does a small change move a lot? What data is missing, how do we get it? By the end we have created a plan we can point at: this is what we're building, this is why, and everyone is on the same page.
I start with quick wins and a working prototype. Could be an optimization model, a simulation, an AI tool, whichever fits the problem. Now it's time to discuss: you see something real, we talk about what works, and set the direction before we start finalizing.
You walk away with tools, dashboards, or a concrete action plan your team can run with. Not a hundred-page report nobody reads. You get documentation and hands-on training where it's needed, so the solution keeps working for you reliably in the long run.
Simplified versions of methods I use in actual projects. They're not meant to be realistic, just enough to show how the ideas work.
Programming, optimization, and AI. Part-time at the University of Hamburg, regular guest lectures at other universities, plus the occasional company workshop for teams. A few teaching awards too, in case that matters.
Start from zero and build real programs. Covers scripting, data structures, and the tools you need to get productive.
Linear, integer, and network models with JuMP. The kind of resource allocation problems that come up in practice.
Structured analysis and quantitative models for managers who want to make decisions based on numbers, not gut feeling.
Something specific in mind? I build workshops from optimization, simulation, AI, and data analysis modules around what your team actually needs.
Plenty more at university. See all lectures at tobiasvlcek.com/lectures
Hands-on work from Hamburg. I handle the core myself and pull specialists from my network when a project needs more.
Photo: Mirjam Kilter
I’m Dr. Tobias Vlćek. Complex operations where off-the-shelf software gives up are my thing. The fun is digging into the specifics, connecting dots in unconventional ways, and building tools that hold up against real-world mess. The real bottleneck sits outside the code: understanding what actually happens on the ground, with the people who live the problem.
I mostly work in logistics, public safety, and transportation. For larger projects I pull specialists through Hamburg Analytics, my network. One person to talk to, a team that actually delivers. It works the other way too: I often join existing teams as their external expert.
For projects at the intersection of science, design, and technology, I partner with scientific illustrator Katharina Vlcek. She brings biology, cultural history, science communication, and visual storytelling; I bring the interactive and technical side. Together we handle museum exhibits, educational materials, and digital product design (this website included).
Reach out, even if you're not sure yet what's needed. Whether it ends up being a web app, a simulation, an optimization model, or ML, we'll figure it out together in a first conversation.