CEU Electronic Theses and Dissertations, 2020
| Author | Domokos, Barna |
|---|---|
| Title | Can Geospatial Data Tell Us Where to Place More Fuel Stations in Hungary? |
| Summary | Markets expand or shrink on longer term. Fuel selling markets behave alike. Companies need to align their strategies in order to counterbalance disruptive forces and explore the emerging new opportunities. There are many attempts to predict how and when the market underlying and shaping forces will open new window for investment both for the existing vendors and for newcomers. Even by analysing competitive vendor's past behaviour and also zooming in on related periods’ sales performance, the right and actionable information or data cannot be easily and quickly grasped and pushed in front of decision makers who are under pressure to take decisions about expansion or investment. For example when deciding about where to place a new fuel station. Domain experts’ advice is also needed to support the idea behind the taken investment decision. In the domain where this capstone project is embedded (fuel selling market, Hungary) we can assume that the growth of a fuel selling company can be ensured in one hand by triggering the revenue stream generators of the business entity (price positioning, offering new services). If the economy goes well. If recession sets in, the first question on the agenda is, how to stay alive, perpetuate the current position on the market and keep the business running. On the other hand one can think in green field investment - such as opening up new shops, divisions in new areas. When a company decides about acquisition or green field investment it is expedient and advisable to assess the purchasing power of that particular zone and calculate with the underlying forces that drives and shapes consumer demand. The question is, how one can reveal and interpret them? There are several methods for this in use. One core element that features all of them: they are based - or at least we assume they are - on real data. Reliable data. Most companies sit on huge piles of data, unstructured and unused data. Many of these companies are not aware the value of the unutilized asset which is called consumers related big data (consumer profile data). In this capstone work I made the attempt to make prediction utilizing new type of data – geospatial data. I framed a simple question: where are white spots, their geospatial location on Hungary’s map where more fuel station could be deployed? Having obtained no real sales data from existing market players to make prediction utilizing quantitative data, I decided to look for independent data sources. Free of charge and freely accessible for anybody. At the end of the project insights were unfolded: by using the methodology learnt at CEU data science courses - namely running regression analysis on multiple independent variables and crafting prediction models and making them to compete – some spots were indicated on the map of Hungary where deploying new fuel stations could be considered. Or at least kept on investors' agenda for further analysis purpose. |
| Supervisor | Eszter Windhager-Pokol |
| Department | Economics MSc |
| Full text | https://www.etd.ceu.edu/2020/domokos_barna.pdf |
Visit the CEU Library.
© 2007-2021, Central European University