
The article presents an exemplary analysis of basic parking measures – the occupancy level of parking spaces in the Paid Parking Zone in Krakow and the turnover of vehicles using it, obtained using a computational algorithm developed by the author and implemented in the Python programming environment. Using multiple linear regression methods, prediction models for occupancy levels and turnover in traffic zones of the central part of the city were created. The explanatory variables of the model were basic demographic data obtained from the transport model of the Krakow agglomeration.
Introduction
In the era of ubiquitous digitalization, so-called Big Data is collected every day, at almost every step. Servers located around the world store numerous traces of activity from mobile phone users, internet users, and store customers. In this way, they constantly provide information that allows for the observation of current sales results or the real-time locations and movements of car drivers. Data owners, on the other hand, are able to better adapt to the dynamically changing situation in the market or on the road. They can process numerous collected data and, based on them, look for causes and analyze the effects of past events, as well as create scenarios and models of future events. Information contained in large databases can be used by numerous private enterprises as well as state and local government administrative units. It is already said today that the use of Big Data constitutes a breakthrough in company management, allowing for the creation of new business models. The emergence of digitized, huge sets of information creates unprecedented conditions for people studying phenomena in many fields. Alongside sectors such as marketing or banking, there is also broadly understood logistics and transport engineering.
Literature review
Acquiring data and the role of its proper interpretation is one of the most significant challenges of modern science, also in the field of transport. For hundreds of years, data was collected and archived in paper form, constituting documentation of events or transactions that took place in the past. Currently, primarily thanks to the digitalization of data, ubiquitous access to the Internet, and sufficient server capacity, the flourishing of technologies related to Big Data analysis has become possible. There are many sources of Big Data in transport applications. They can be city bike systems, registers of delivered courier parcels, or taxi corporation applications. An example of a Big Data application that can be found in the literature is, among others, supply chain management. Based on the analysis of large data, companies are able to optimize routes on an ongoing basis, plan operational capacities in detail, perform comprehensive risk analyses, etc. Another way of using Big Data for broadly understood transport analyses is also the creation of origin-destination matrices based on data from mobile operators. These matrices are one of the possible sources of input data for transport models. These models can then be calibrated, at subsequent stages of work, to speed measurements obtained from massively collected GPS data. Another example of a transport application of Big Data is attempts to predict (in real-time) the occupancy levels of publicly available parking spaces. Data from one of the mobile phone operators, through which it was possible to pay for parking, was used for the analysis. The above-mentioned examples, taken from the literature, allowed for the identification of diverse applications of Big Data in transport issues. However, they constitute only a small percentage of the possibilities offered by so-called big data. Both the scope and the subject matter of the analysis are, when possessing large data sets, limited primarily by the constraints of human perception and imagination.
The full article is available below in PDF format.









