University of Twente Student Theses
Forecasting spatial and temporal variations in OD-pairs : a case study from Sao Paulo
Engels, J.M. (2019) Forecasting spatial and temporal variations in OD-pairs : a case study from Sao Paulo.
|
PDF
185MB |
| Abstract: | The recent years have seen many developments in the field of sustainable transport. One of these trends is the deployment and development of public bike sharing systems in various forms. These bike sharing systems provide the customers with bicycles on as needed basis. Since 2012 such a system is available in Sao Paulo, Brazil. The system, that is analyzed in this research, allows the customers to pick up the bicycles from fixed stations. The system that is in operation in Sao Paulo consists of 261 of these fixed stations. The first system that can be considered a bike sharing system was deployed in 1965 in Amsterdam, the Netherlands. Since then the systems and especially the bikes used within the system have developed. They improved from simple unlocked white painted bikes to sturdy bikes with online locking mechanism and GPS-tracker. Since the middle of the 1990’s researchers have started investigating and analyzing these systems. Since this period the number of systems that operate in different cities across the world also increased. As the number of systems worldwide increased they gained the attention of the academic society. There are many different researches that analyze various aspects of the bike sharing systems. From a city planner point of view researchers identified strategies on how to integrate these system into the existing public transport system. Another aspect that has been studied with respect to PBSS is for what trip purposes the bicycles are used. For the operator of the systems it is very interesting how the bicycles move within the system and between stations. The topic that has been studied the most extensively, are the factors that influence the trip attraction and generation of each station. The factors that influence the number of arriving and departing trips per station and that have been identified by the literature are; surrounding available bicycle infrastructure, land use around the station, proximity to other transport modes, environmental aspects (slopes and weather) and population characteristics. With respect to the available infrastructure many researches have shown that the quality and presence can positively influence the number of trips (e.g. [Cleland and Walton, 2004]). When it comes to the land use, a more divers land use around a station can increase the number of arriving and departing trips per station (e.g. [McBain and Caulfield, 2017]). The distance to connecting public transport plays an important role for the number of trips per station. If the distance to other public transport modes from a station decreases the number of trips increases [Raux et al., 2017]. The city characteristics like topography and size have been identified by the research of [Cleland and Walton, 2004] to have an influence on the number of trips. The distance from home or work to a station is another factor that influences the number of trips per station [Wang and Lindsey, 2019]. Next to identifying the factors that influence the number of trips per station, researchers also tried to forecast the number of arriving/ departing trips or the number of available bikes per station for given moments in time. For these forecasts different approaches like Markov-chains or ARIMA-models have been used. Until now there is limited number of researches that focused on forecasting the number of trips between stations, so called OD-pairs and therefore forecasting spatial variations in the number of trips. It is unknown if the factors at the station level, that have been identified by other works, apply to the case of Sao Paulo. Based on the gaps in the literature the following research objective was defined: Develop a model that can forecast spatial and temporal variations in the number of trips between OD-pairs in a PBSS. First of all the factors that have been identified for other systems will be tested for the case of Sao Paulo. It is unknown if the factors that influence the number of arrivals/departures at station level also apply for OD-pairs. Therefore it will be tested if the factors that apply to stations in the PBSS of Sao Paulo are also suitable to describe spatial variations. The unique thing of this research is that the research goes beyond just identifying the factors that influence the usage at the station level. This approach is possible due to the amount and quality of the available data. This research will use different data-sources to identify the factors. For the surrounding characteristics of each station information from the municipality and the LABGEO will be used. The individual trip data, the most important information for this research, was provided by the Brazilian center for analysis and planning (CEBRAP). For this research the historical data of rental processes from 2012 until 2017 has been used to identify the number of trips. This researches includes 261 stations and 438.862 individual trips. 4 |
| Item Type: | Essay (Master) |
| Faculty: | ET: Engineering Technology |
| Subject: | 56 civil engineering |
| Programme: | Civil Engineering and Management MSc (60026) |
| Link to this item: | https://purl.utwente.nl/essays/77460 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page