Infrastructure Engineering - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemToward Constant Service Quality Monitoring in Transport Nodes( 2022)Promoting public transport relies on developing effective tools for the proactive management of service quality in public transport nodes. However, the challenging environment of these nodes, characterised by crowding and congestion, limited physical space, and frequent changes, poses obstacles to existing monitoring approaches. This calls for novel data-driven methods to provide insight into the management of such a challenging environment. This study fills this gap by combining subjective and objective service quality measures, along with proposing state-of-the-art models, to enable constant assessment of service quality in the confined spaces of transport nodes. The resulting insights are expected to support prompt, context-aware, and flexible decision-making, benefiting local communities, transport authorities, and related industries. First, the feasibility of integrating unstructured data sources, such as social media content, for concurrent monitoring of service quality in transport nodes is investigated. To that end, tweets are mined to evaluate various service quality characteristics. To realize this, a novel framework based on a fine-tuned language model and sentiment analysis is proposed. This allows for the classification of tweets and the effective detection of unusual events impacting perceived service quality. Findings highlight the constraints of sparse yet valuable data, where extra knowledge from sentiment analysis improves the monitoring tool's sensitivity in a variety of settings. Next, a novel framework is designed to discover and predict events impacting service quality across a network of constrained public transport nodes. To that end, social media content offers a unique opportunity to effectively model such a complex phenomenon. Nonetheless, this is challenging. Apart from data sparsity, the asynchronous nature of the observations impedes pattern discovery and event prediction. To tackle this, Hawkes Point Process is used to model events without making them discrete. Moreover, sentiment analysis is used to strengthen the model with extra information. Results demonstrate the effectiveness of the approach in identifying the causal patterns and predicting events over the limited context. Then, Closed-Circuit Television (CCTV) footage as an authoritative dataset is used to monitor pedestrians' behaviour on platforms and objectively measure service quality in transport nodes. A pre-requisite is to accurately identify and localize pedestrians in the footage. However, this is challenging as pedestrians are often partially occluded in the crowded environment of transport nodes. To address this, a novel human pose-aware pedestrian localization framework is proposed to ensure a fast and accurate location of pedestrians. Results show the effectiveness of the approach in pedestrian localization. Finally, a novel framework for an effective combination of subjective and objective service quality measures in transport nodes is provided. To this end, multivariate Hidden Markov Models and vision-based Level of Service estimation are used to ensure robustness in the fusion of both datasets with significantly different characteristics. The model is strengthened by employing extracted sentiment information from social media feeds as additional knowledge about passengers' perceptions. Results show the efficacy of the framework compared to data-intensive state-of-the-art data fusion approaches. Overall, this study develops a range of new methods and brings new theories and technologies together, to enable concurrent monitoring of public transport infrastructure performance. The results of this study can be used as a contributing module in Cyber-Physical systems to constantly monitor passenger-infrastructure interactions. The findings would not only have a positive impact on people's quality of life but also inspire researchers from other disciplines, especially the public transport community, to employ data-driven solutions in a variety of interesting ways.
-
ItemDevelopment of efficient flood inundation modelling schemes using deep learning( 2022)Flood inundation models are one of the important tools used to manage flood-related risks in engineering practices such as infrastructure design, flooding disaster mitigation, and reservoir operations. Two-dimensional (2D) hydrodynamic models are commonly used in engineering applications because of their ability to provide robust estimates of flood inundation depth and extent at high temporal and spatial resolutions. However, due to the high computational costs, 2D models are not suited to many applications such as real-time ensemble flood inundation forecasting or uncertainty analysis. Therefore, many models have been developed based on simplified hydraulic rules such as considering only the conservation of water mass. These models are generally faster than 2D models but have reduced accuracy, which is a problem in many studies where a fine simulation timestep is needed or flow dynamics are significant. Recently, emulation models have been developed for fast flood inundation modelling using data-driven techniques including artificial neural networks, machine learning classification models, and deep learning. These computationally efficient emulation models are found to have comparable accuracy with 2D models when used to simulate flood inundation water level or depth provided with rainfall or streamflow discharge inputs. However, most emulation models simulate flood water/depth for each grid cell in the modelling domain separately, which would significantly increase the computational costs when applied for large domains. To add to that, these models have been found to have reduced accuracy in data-scarce regions on the floodplain. To improve the performance of emulation models, the objective of this thesis is to develop computationally efficient flood inundation models using deep learning and new spatial representation methods, that can be used for fast flood inundation simulation on floodplains with various characteristics at high spatial and temporal resolutions. The major contributions of this thesis include: (1) the development of an emulator for rapid flood inundation modelling which incorporates a novel spatial reduction and reconstruction (SRR) method as well as long short-term memory (LSTM) deep learning models to efficiently estimate flood inundation depth and extent; (2) the development of a Python program for the SRR method for flood surface representation; (3) the development of a U-Net-based spatial reduction and reconstruction (USRR) method and one-dimensional convolutional neural network (1D-CNN) models to emulate flood inundation on flat and complex floodplains. In addition, an input selection structure is developed and validated in the architecture of the LSTM models to simplify the model development process and to reduce the effort required for real-world applications. Also, a comparison is carried out for the performance of the combined approaches of the SRR method and LSTM models, as well as the USRR method and 1D-CNN models in an application to a flat and complex floodplain. The comparison demonstrates the advantages of using the USRR-1D-CNN emulator for rapid modelling of flood inundation on flat floodplains with complex flow paths, while the SRR-LSTM emulator is more computationally efficient and suitable for application to steep floodplains. The flood inundation modelling schemes developed in this thesis provide fast estimates of flood inundation surfaces without a material loss of accuracy compared to 2D hydrodynamic models, useful for applications such as ensemble real-time flood forecasting and flood risk analysis. They have the potential to deepen our understanding of the impacts of input uncertainty on temporal and spatial patterns of flood inundation, and to facilitate improved flood risk management.