Every student their own subjects
In the upper years every student chooses their own subjects. Electives have to be grouped into clusters and students assigned to them, before a single lesson is scheduled.
Jomin takes a fresh look at school timetabling software, built on two principles: exceptionally good optimisation and exceptionally simple software.
A school timetable has to work for hundreds of students, dozens of teachers and a limited number of rooms, all at once. Mathematically it is among the hardest planning problems there are: every extra rule makes the puzzle considerably harder.
In the upper years every student chooses their own subjects. Electives have to be grouped into clusters and students assigned to them, before a single lesson is scheduled.
Every lesson needs a teacher, a room and a time when everyone is available. After that, free periods, spread and the wishes of teachers determine how well it works in practice. Which of these matters most differs from school to school.
A teacher's hours change, a group is split, a room drops out. Then you want to adjust without turning everything upside down.
Exceptionally good optimisation under the hood, and software that feels as simple as possible to the people who use it.
We are building a modern optimisation engine that computes complex school timetables, respecting every hard constraint and the varied wishes of students, teachers and schools. We start with the base timetable, including clustering and student assignment; later the same engine should also be able to adjust an existing timetable, with as little disruption as possible.
We don't just want to produce timetables; we want to find out how good they can actually get. So we test the engine on well-known public timetabling problems, alongside existing solutions and at equal computing time, and we check every result independently.
And we make it visible why a choice was made and which trade-offs are possible.
Building timetables is complex work. We want the software not to be: an intuitive interface that keeps the complexity behind the scenes wherever possible. Where timetablers need that simplicity most is something we test in practice.
Simplicity is not a limit on functionality, but a design principle.
We don't want to build a product on assumptions about how schools work. So we start with the timetabler: conversations, observing at schools, studying existing solutions and collecting realistic timetabling problems.
We want to learn the timetabler's language, understand the implicit trade-offs involved and find out where the real bottlenecks are.
Jomin is still in development. The user experience, the features and the eventual shape of the product will follow from what we learn in practice.
To combine the intelligence of advanced optimisation with the simplicity of carefully designed software.
Not by building as many features as possible, but by solving the right problems exceptionally well.
We build software that supports timetablers in their craft, gives schools more insight and flexibility, and ultimately leads to better timetables for everyone.
Timetabler, school leader, software vendor or just curious? We'd love to hear from you.