Tools
Tools
-, Pragya Jha (2024) Enhancing the Dublin Bikes System: An Impact Analysis to Identify Pre, During, and Post-Pandemic Trend. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (2MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (1MB) | Preview |
Abstract
The sudden attack of COVID-19 outlined the need for a resilient and adaptable public transportation system. Since the pandemic impeded the dynamics of urban mobility, causing overcrowding and thus lockdowns the systems traditionally operating were not able to put up with the challenge showing that more flexible transport alternatives are called for. There is extensive research on resilience within public transportation systems; however, the contribution of bike-sharing systems in maintaining urban mobility during such crises remains underexplored. This paper is a detailed analysis of the Dublin bike-sharing system given its potential to be an effective mode of public transport when there might arise a repeat pandemic or events of that nature. The critical analysis, extending up to data received in 2018 to 2024, proposed the use of variants of predictive models, namely Random Forest Regressor, Linear Regression, ARIMA, LSTM, SARIMA, and Prophet for bike availability and usage pattern prediction. To arrive at this methodology, exhaustive data cleaning and processing had been done to offer the expected and accurate level. After such data cleansing, each model was trained and tested to estimate their predictive performance. Of these, the Random Forest model proved the most accurate, with an R-squared value of 0.93. The findings are indicative of the fact that utilization patterns of the bikes changed in crucial ways during the pandemic, in the form of increased demand and peak shifting. The results indicated that the bike-sharing system in Dublin played a fundamental role in replacing demand during a time when the traditional transport systems were choked. The author has suggested several recommendations based on findings that would make the system resilient: dynamical reassignment of bikes by evolving demand, developing infrastructure that would flex or adapt to new circumstances quickly, ramping up sanitization protocols to keep the public safe and clearer messaging with customers about bike supply and ways to maintain health. The paper also identifies some possible future directions of research by which, through the integration of health data and collaboration with public health authorities, further improvement in system adaptability and resilience may be achieved. Such measures would make the bike-sharing system of Dublin more efficient but also substantially better prepared to serve as a reliable means of public transport in any future public health crisis.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Rustam, Furqan UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HT Communities. Classes. Races > Urban Sociology > City Planning R Medicine > Diseases > Outbreaks of disease > Epidemics > COVID-19 Pandemic, 2020- Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 06 Aug 2025 14:22 |
| Last Modified: | 06 Aug 2025 14:22 |
| URI: | https://norma.ncirl.ie/id/eprint/8445 |
Actions (login required)
 |
View Item |