Planning optimization made easy.

Many optimization solvers are designed for academic or isolated problems, such as the Traveling Salesman Problem, VRP, Bin Packing or Job Shop Scheduling.
In reality, planning problems are more complex. They combine hard/soft constraints from different use cases and often include country or company specific constraints.

Timefold was created to automate real world planning optimization with ease. Solve for hard, soft and other constraints. Reduce costs substantially, improve service quality, fulfill employee wishes and lower carbon emissions. Far more than expected.

Traditionally, optimization solvers are the domain of mathematicians and expensive, specialized Operations Research consultants. You feed those solvers your data as number matricesand your constraints as mathematical equations.

Timefold is build for programmers. You feed it your data as domain classes and define your constraints as code. Let us worry about the math inside the solver. Your code becomes self-explanatory. Integration with other APIs is straightforward. And when the business rules change constraints - they eventually always do - maintenance is easy.

This is a revolution in the world of planning optimization, and a central element in the mission of our company, to increase the accessibility of planning optimization.

Timefold Community is Open Source software, released under the Apache License 2.0. Use it in your commercial software, for free. It's a complete and professional solver. We frequently release new features, enhancements and fixes through Maven Central (Java, Kotlin) and Pypi (Python).

Our entire company believes in the power of open source to build high-quality software. Our team participates in the open source community and regularly contributes to other open source projects.

To pay for Timefold Community development, we sell Timefold Enterprise: a Source Available extension with high-scalability features and enterprise support.

Timefold is extremely fast. It combines performance tricks from metaheuristic algorithms, concurrent programming, database indexes and incremental (delta) calculations with native compilation support.

But speed without scaling is irrelevant. The search space of planning problems scales exponentially to the size of the problem. The quality of a solver on a small dataset says nothing about that solver on a big dataset.

Timefold is arguably the most scalable production solver on this planet. Like other solvers, it's CPU-bound with little to no I/O. Unlike other solvers, memory consumption barely increases as you scale out, making it ideal for cloud deployments.

Timefold is build for operational planning, not just strategic or tactical planning. We deeply understand that a schedule is living, breathing thing. That its requirements change over time. That planning agility is essential for great results. Therefore, Timefold handles:

• Continuous planning: Publish a schedule every week, several weeks before execution.
• Pinning: Your user is still in control. Timefold plans around their manual assignments.
• Non-disruptive replanning: Deal with changing circumstances, but minimize changes to the published schedule.
• Overconstrained planning: When there are too few resources to cover all the work, help find a solution.
• Real-time planning: React on real-time disruptions of the plan within milliseconds.