2026-04-13T02:32:51Zhttps://recercat.cat/oai/request

oai:recercat.cat:2117/1031632025-07-23T00:13:01Zcom_2072_1033col_2072_452951

urn:hdl:2117/103163 An analysis of Bicing mobility patterns using big data Manchón Contreras, Oriol Àrees temàtiques de la UPC::Enginyeria civil Big data Bicycle commuting big data bicing mobility patterns Macrodades Desplaçaments en bicicleta Nowadays, technology advances really fast and so does the generation of data. Almost all electronic devices are constantly generating and sharing a huge amount of data through the World Wide Web. Moreover, recent policies of open governments and data, are helping to make available this information for everybody that wants to take it and use it. The aim of using Big Data is to discover knowledge that is hidden behind thousands of rows of information. However, to find out the value of the data, it is necessary to use non-traditional methods able to deal with such amount of information. Furthermore, big cities have traffic problems and complex mobility patterns which need to be studied in depth to improve life conditions of citizens, reduce pollution and to create eco-friendly cities. This work is focused on the city of Barcelona and its bike-sharing system Bicing. The aim is to understand the mobility patterns of Bicing subscribers using Big Data. Treating Big Data requires of more resources than conventional problems. So that, setting a methodology to acquire, pre-process and treat the data has been necessary before proceeding with the analysis. In order to gain visibility out of the data, two different approaches have been followed. First of all, an exploratory analysis of the behaviour of the users of Bicing. On the other hand, a Principal Component Analysis has also been carried out to understand the data but also to reduce the dimensionality, hence the volume of the data necessary to provide acceptable results. To sum up, the present work is a particular example of the possibilities that Big Data offers in terms of gaining knowledge out of massive amounts of data. Moreover, it studies the patterns of Bicing subscribers during different periods of the day, week and year based on real data. 2016-06-23 Master thesis http://creativecommons.org/licenses/by/3.0/es/ Open Access Attribution 3.0 Spain Universitat Politècnica de Catalunya