|
|
Application for searching connections in public transport systems with the use of machine learning
Paulík, Kryštof ; Hynek, Jiří (referee) ; Lazúr, Juraj (advisor)
With the growing population of cities, there is an increase in the number of cars as well as other means of transport in the streets, leading to traffic complications causing delays in public transit vehicles. Users planning their public transit journey need to simply search for the fastest route. However, this gets complicated by vehicle delays, which can often lead to missed transfers or other problems. The intention of this thesis is to introduce machine learning to public transit route search so that possible delays of vehicles are taken into account when searching for a route. The application aims to reduce the time passengers spend on their journey by finding more reliable routes in the public transit system.
|
|
|
Analysis and Visualizaion of Brno Public Transport Data
Lazúr, Juraj ; Škarvada, Libor (referee) ; Hynek, Jiří (advisor)
The planning and management of modern public transport systems requires knowledge of the behaviour of the systems in a real environment. The aim of this thesis is to simplify the procedure of extracting useful knowledge from vehicle delay data in the Brno public transport system. The solution to the problem consisted in the design and implementation of a tool that will autonomously process and analyze this data. The outputs generated by the implemented tool provide support for planning, implementation and evaluation of changes in the public transport system. They also allow the detection of critical points and recurring patterns in the system. The results of this work allow for a more efficient operation of the public transport system and should serve to increase the satisfaction of the travelling public.
|
|
| |
|
|
Analysis and Visualizaion of Brno Public Transport Data
Lazúr, Juraj ; Škarvada, Libor (referee) ; Hynek, Jiří (advisor)
The planning and management of modern public transport systems requires knowledge of the behaviour of the systems in a real environment. The aim of this thesis is to simplify the procedure of extracting useful knowledge from vehicle delay data in the Brno public transport system. The solution to the problem consisted in the design and implementation of a tool that will autonomously process and analyze this data. The outputs generated by the implemented tool provide support for planning, implementation and evaluation of changes in the public transport system. They also allow the detection of critical points and recurring patterns in the system. The results of this work allow for a more efficient operation of the public transport system and should serve to increase the satisfaction of the travelling public.
|
|
|
Off-line connection search on Google Android platform
Křepelka, Michal ; Bojar, Ondřej (advisor) ; Straka, Milan (referee)
This thesis discusses the connection search in public transit without permanent connection to the server, that would do the time-consuming calculations. For this purpose, we use the Transfer Pattern method running on Google Android platform. We demonstrate to the reader some of the most common graphs used for connection search in public transit and subsequently the procedure how to formalize timetables as such graphs. Further we describe principles of Transfer Patterns, a way how to compute them from timetable graph and how to store them in SQLite database on Android device. On such pre-computed data, we can very quickly and efficiently find the optimal connection even on relatively performance-limited Android device.
|
guest ::
