Dopravní Data Predikce

Denis Varaksin

Master's thesis

Traffic Data Prediction

Abstract:

Data analysis and data prediction is the field of informatics and mathematics, en-gaged in calculation of algorithms and mathematical models that are able to extract practical data from analyzed data. Data analysis has many aspects and approaches, covers different methods in various fields of science and everyday human life. Data prediction and forecasting has interested people for thousands of years, with the new stage of human civilization development - expenditure of computers and different computing machines, data prediction methods and techniques tremendously change. New field of "Big Data" and machine learning, which research data sets that are too large to deal with by traditional data analysis techniques and applications are expanding. In our days "Big Data" are widely used in areas of internet search, economics, business, urban informatics and etc. The urban informatics is one of the most interesting and applicable fields of "Big Data" usage. This field uses information and data sets in the context of smart cities and urban environments with purpose to make quality of life of pedestrians better and improve urban environment. The aim of this project is to create a model, which would predict behavior of one of the most visible part of every urban area - crossroad. Provided information from traffic light controllers (detectors) on the crossroad at "Makro" Zlin is being registered, stored with equal periods of time and analyzed. Data analysis is implemented through usage of different statistical and computation models in a free and open-source integrated development envi-ronment "RStudio" and spreadsheet program for data storage "Microsoft Excel". The project is aimed to predict traffic data on the crossroad.

Abstract:

Data analysis and data prediction is the field of informatics and mathematics, en-gaged in calculation of algorithms and mathematical models that are able to extract practical data from analyzed data. Data analysis has many aspects and approaches, covers different methods in various fields of science and everyday human life. Data prediction and forecasting has interested people for thousands of years, with the new stage of human civilization development - expenditure of computers and different computing machines, data prediction methods and techniques tremendously change. New field of "Big Data" and machine learning, which research data sets that are too large to deal with by traditional data analysis techniques and applications are expanding. In our days "Big Data" are widely used in areas of internet search, economics, business, urban informatics and etc. The urban informatics is one of the most interesting and applicable fields of "Big Data" usage. This field uses information and data sets in the context of smart cities and urban environments with purpose to make quality of life of pedestrians better and improve urban environment. The aim of this project is to create a model, which would predict behavior of one of the most visible part of every urban area - crossroad. Provided information from traffic light controllers (detectors) on the crossroad at "Makro" Zlin is being registered, stored with equal periods of time and analyzed. Data analysis is implemented through usage of different statistical and computation models in a free and open-source integrated development envi-ronment "RStudio" and spreadsheet program for data storage "Microsoft Excel". The project is aimed to predict traffic data on the crossroad.

Keywords

Traffic, data, prediction, analysis, ARIMA, forecast, AR, MA, Loess, Crossroad

Language used: English

Date on which the thesis was submitted / produced: 17. 5. 2019

Thesis defence

• Supervisor: Ing. Dušan Hrabec, PhD.

Tomas Bata University in Zlín

Faculty of Applied Informatics

Master programme / field:
Engineering Informatics / Information Technologies