How Vilayat works

What it does

You pick a university campus, the most you can pay in rent and the longest you are willing to travel each morning. Vilayat then ranks the city's neighbourhoods by how well they fit, shows them on a map and explains each position in plain words.

It ranks areas, not flats. When you have chosen a few areas, it links you to the usual listing portals so you can search there. Vilayat does not copy listings from those portals and does not show vacancies.

The three views

• Planner: the map and the ranked list. You can also type what you want as a sentence, for example “near TUM, under 800 euros, quiet”, and it fills in the settings for you.
• Compare: the same budget at every university in the city, to see where it goes furthest.
• Ask: questions about registration, the residence permit, health insurance, working while studying and similar paperwork, answered from a set of stored documents with the source named.

How the ranking is worked out

First, areas that break one of your hard limits are set aside: rent above your ceiling or a typical commute longer than your limit. They are still listed, with the reason they were excluded.

The remaining areas get a score between 0 and 1 for each of four things, and a weighted average decides the order:

• Commute (35% by default): 1 on the doorstep of the campus, 0 at your own limit.
• Rent (35%): 0 at your ceiling, 1 at half of it or less.
• Everyday amenities (15%): supermarkets, cafés, libraries, gyms and similar places per square kilometre.
• Part-time job access (15%): how many places that typically hire students are within reach. This shows where such employers cluster; it is not a count of open jobs.

You can move the weights yourself, or mark a priority as important, which raises its weight above all the defaults. Every reason shown next to an area is written by the same code that scored it, so the explanation always matches the number.

How commute times are measured

The city is divided into small hexagons, about 0.7 km² each. For every hexagon, the travel time to each campus is computed by public transport and walking, using the transport operator's published timetable and OpenStreetMap footpaths, at walking speed 4.8 km/h. The trip is tried for every departure minute between 07:30 and 09:00, and the middle value (the median) is kept, so one lucky or unlucky connection does not decide the figure.

An area's commute is the median over its hexagons. The commute layer on the planner map shows the hexagons themselves, because the far edge of a large district can be much further from the campus than its centre. Two neighbouring areas can therefore show the same figure, or quite different ones if a fast line runs through only one of them.

When you open an area, the planner also shows one typical trip step by step (walk to the stop, which line, where to change, walk to campus) and draws it on the map. It starts from the hexagon whose time is closest to the area's median, and of the trips leaving between 07:30 and 09:00 it is the one whose length is closest to that hexagon's typical time. It shows how the trip goes, not the fastest possible connection.

All of this is computed in advance, so searches answer straight away. Where Vilayat has no figure for something, it says “unknown” rather than guessing.

Where the data comes from

Each dataset, its publisher, licence and date are listed on Data and sources, which also explains the figures marked with an asterisk. How the language model is used, and what it is not allowed to decide, is on How we use AI.

Who made it

Vilayat is a student project, built as a B.Tech minor project on generative AI and cloud computing (project name EduCommute Matrix). It is not run by a university, a city or a housing company.