Analiza prometnih tokova autoceste Zagreb-Karlovac primjenom neuronskih mreža

master's thesis

• Title (english): Analysis of Traffic Flows at Zagreb-Karlovac Motorway using Neural Networks
• Author: Marko Slavulj
• Mentor: Edouard Ivanjko (mentor)
• Committee member: Marko Šoštarić (predsjednik povjerenstva)
• Committee member: Martin Gregurić (član povjerenstva)
• Committee member: Tonči Carić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Transport and Traffic Sciences; (Division of Intelligent Transport Systems and Logistics); (Department of Intelligent Transport Systems); Zagreb
• Defense date and country: 2021-09-22, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Traffic and Transport Technology; Intelligent Transport Systems and Logistics

Abstract

Sa sve većom prometnom potražnjom na autocestama, osnovni prometni pokazatelji razine uslužnosti kao što su: sigurnost, vrijeme putovanja i onečišćenje okoliša dolaze do sve većeg izražaja i u našoj zemlji. To je posljedica sve većeg obujma turista koji svoj godišnji odmor upravo provode na teritoriju Republike Hrvatske, ponajviše na obali Jadrana. To dovodi do potencijalne opasnosti na autocestama poput prometnih nezgoda i nesreća, nezadovoljstva korisnika dobivenom uslugom, itd. Republika Hrvatska ima jedne od najrazvijenijih autocesta u Europi pa i time prikuplja veliku količinu prometnih podataka. Ti prometni podatci osnova su za procjenu stanja prometnog sustava i prijedloga mogućih rješenja. Prometni podatci običnom čovjeku ne predstavljaju ništa dok se ne obrade i analiziraju. Za rješenje problema obrade velike količine prometnih podataka primijenjena je metoda K-srednjih vrijednosti (engl. K-means) koja određuje optimalni broj grupa koje sadrže dane u godini te je primijenjena samoorganizirajuća neuronska mreža koja analizom prikazuje potencijalne van-distribucijske raspodjele i grupe podataka. Također je analizirana i vizualizirana kategorizacija vozila, brzine i sezonalnost prometa.

Abstract (english)

With increasing traffic on motorways, basic traffic level of service indicators, e.g.: safety, travel time and environmental pollution, are coming to the fore. This is a consequence of a growing number of tourists who spend their vacations in Croatia, mostly on the Adriatic coast. This again leads to potential risks on motorways, such as car accidents and collisions, customer dissatisfaction, etc. Republic of Croatia has one of the most developed motorway networks in Europe, and it collects a large quantity of data. These traffic data is basis for checking the conditions of the traffic system and proposing possible solutions. To an average person the traffic data does not mean anything, until it is processed and analyzed. To solve the issue of big traffic data processing, K-means clustering was applied as a method of processing traffic data. It determines an optimal number of clusters, that contain days of the year, then a self-organizing neural network is applied, which, through analysis, shows potential outlier allocations and clusters. On top of that, categories of vehicles, speed and seasonality of the traffic is analyzed and visualized.

• Keywords: k-means; neuronske mreže; prometni tok autoceste; kategorizacija vozila; samoorganizirajuća neuronska mreža (SOM)
• Keywords (english): k-means clustering; neural networks; highway traffic flow; vehicle categorization; self-organizing map (SOM)
• Language: croatian
• URN:NBN: urn:nbn:hr:119:687508
• Study programme: Title: Intelligent Transport Systems and Logistics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka prometa (magistar/magistra inženjer/inženjerka prometa)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2021-10-25 15:14:12