Infrastructure Engineering - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 12
-
ItemLightweight traffic anomaly detection: A case study with SCATS volume data of Melbourne(Australasian Transport Research Forum, 2023)In this paper, we evaluate performance of an anomaly detection framework with real traffic count data collected by SCATS (Sydney Coordinated Adaptive Traffic System) loop detectors in Melbourne. The goal is to detect anomalous daily volume profiles within temporally large historical traffic data utilizing a lightweight and parameter-free approach and use it for live applications. To achieve this, daily volume profiles are first compressed into two dimensions benefiting from the Principal Component analysis (PCA). Then, a parameter-free version of DBSCAN is applied to the data with unique days of the week. Results from more than 20 different locations in Melbourne are fully visualized and the advantages and disadvantages of the method are discussed. We found that, with this approach, anomalous volume profiles can be accurately detected in a wide range of spatiotemporal data without any pre-training, parameter setting, or using complex learning methods.
-
ItemTesting request prioritization strategies to improve the quality of a shared autonomous vehicles service: A Melbourne case study(Australasian Transport Research Forum, 2023)Shared autonomous vehicles (SAVs) have the potential to revolutionize urban transport by offering a mobility service that combines the benefits of autonomous vehicles and ride-hailing systems. However, one of the main limitations of these services is the handling of all request types in a singular way, which can increase biases towards a fixed type of passengers who have different properties and accessibility-based constraints. This study proposes a prioritization approach based on different request properties and the urgency of the request. The study uses the MATSim simulator to evaluate the prioritization schemes and implements five different scenarios to assess the collective and individual effects of the schemes. The case study focuses on Melbourne Metropolitan Area, and the prioritization is done based on the existing public transport (PT) service and pre-calculated SAV demand of the network. The study considers different performance measures, such as service efficiency, externalities, and provision equity, to estimate the benefits of the prioritization. The results indicate that prioritization can improve overall equity by spreading wait times evenly across the network. The prioritization approach improves the service with more riders finding the service attractive, resulting in more served rides with lower average vehicle kilometers traveled per ride. In addition, the study shows that the PT mode share is increased in multiple scenarios, demonstrating the positive effect of considering accessibility when prioritizing requests.
-
ItemWhat influences passenger’s arrival rate at stops in Melbourne: WASEA-LSTM: a novel deep learning model leveraging multi-source data fusion(ATRF, 2023)Public transportation demand plays a crucial role in service planning and operation. Accurate prediction of passenger arrival rates at transit stops allows transportation planners and operators to optimize resources and improve service efficiency. Current methodologies primarily focus on weather's impact in the aviation industry, supply dynamics, and arrival time prediction, while overlooking its influence on public transport demand variation. This study addresses these gaps by designing a deep neural network model that can predict public transit demand, using large-scale datasets from multiple sources in Melbourne, Australia. We propose a novel deep learning architecture called Wasea-Lstm (Weather-Aware Smart Exponential Activation LSTM) that captures spatial, temporal, and external correlations for passenger arrival rate prediction at tram stops. The model is trained and tested on integrated datasets from automatic fare collection (AFC), automatic passenger count (APC), and weather data over a period of three months. Results show that the Wasea-Lstm model significantly outperforms benchmark models, including gradient boosting machine (GBMR) and multi-layer perceptron (MLP) regression by 15% and 6% in R2 metric, respectively. The feature importance ranking reveals that stop location, time of the day, temperature, and humidity are the key influencers of passenger arrival behaviour in Melbourne. Overall, this study contributes to the development of a model that accounts for multi-dimensional, high-resolution determinants of passenger demand using large-scale datasets from real world. The proposed Wasea-Lstm architecture shows exceptional performance in precisely forecasting stop-level demand for one of Melbourne's largest tram routes. Moreover, its applicability extends seamlessly to all routes within the network.
-
ItemNo Preview AvailableValue of Intersection Cross Box Data in Traffic Signal Control(ITS World Congress, 2022)
-
ItemNo Preview AvailableInvestigating into public transport fare noninteractions using large-scale automatically collected data(Australasian Transport Research Forum, 2022)Fare card data provides an unprecedented opportunity to monitor day-to-day variability of travel demand and its responses to service disruptions and special events. However, when passengers take public transport without interacting with the fare system, demand is usually underestimated, which may cause problems for performance measurement and revenue collection. This research aims to investigate the fare noninteractions phenomena of the tram network in Melbourne, Australia. According to a prior evaluation, only 37% of boarding passengers validate tickets. This study utilizes large-scale automatically collated data to measure fare noninteractions, including data collected by Automatic Passenger Counting (APC) and Automated Fare Collection (AFC) systems. Compared to previous studies with small samples of on-board surveys, it contributes to the state of the art as these high coverage data enable the study of the impact of different types of explanatory variables, including time periods, routes, stop location, travel demand variability, presence of an inspector on-board, etc. Moreover, a free service zone is located in Melbourne central business district where passengers are not required to validate tickets. We specifically investigate passengers’ behavior at the boundary of a free service zone. Results show that fare noninteractions are lower for stops close to train stations, education facilities, stops that have been frequently inspected, and during the peak hours, but are higher for stops with large boarding flows, crowded services, evening periods and weekends. Importantly, conditioning on other variables, fare noninteractions at the boundary of the free service zone are higher in the morning peak but lower in the afternoon peak. The passenger flow diagram demonstrates the reason behind this may lie in the differences between purposes of trips. This investigation provides a starting point for proposing solutions to deal with the missing AFC data due to fare noninteractions.
-
ItemNo Preview AvailableConsiderations for Expediting Road Safety Benefits with the Deployment of Vehicle Communication Technologies(Australasian Transport Research Forum, 2022)Communication technologies are enabling the introduction of connected vehicles and have the potential to improve road safety outcomes at a global scale. This paper aims to deliver a systematic understanding, classification, and evaluation of available communication technologies for road safety that considers the current challenges, mindsets, and future direction for C-ITS technology implementation. This is achieved by combining the results of three lines of research inquiry: 1) literature review of existing communication technologies and worldwide pilot experiments and trial implementations, 2) assessment of the potential for selected connected vehicle safety applications to address motor vehicle crashes across different geographies and road conditions, and 3) expert panel interviews to investigate the challenges and opportunities for technology implementation, specifically in the Australian context, with supporting evidence from global literature sources. These investigations found that C-ITS deployment concerns identified by stakeholders are in line with those identified in literature; however, there are significant safety benefits to be reaped from C-ITS deployment. Policymakers can leverage the potential of this positive outcome and target efforts at addressing the identified challenges when considering pathways to the uptake of connectivity technologies.
-
ItemNo Preview AvailableData-driven evaluation model of safety risks at signalised intersection(Australasian Transport Research Forum, 2022)Near future safety risk evaluation is a critical step towards adaptive traffic safe operation at a smart intersection. This paper proposes a data-driven model that can quickly evaluate simulated safety risks for use in adaptive operational interventions. A traffic micro-simulation model was utilised to generate conflicts-based data for developing the machine learning model. Conflict indicators including time to collision, TTC, and post encroachment time, PET, were used to identify safety risk. Supervise learning models such as linear regression and machine learning models including random forest and extreme Gradient Boosting (XGBoost) were employed to evaluate risk indices for adaptive operations. In total, 9 models were trained, and XGBoost were found to outperform the other algorithms with 0.87 of the overall accuracy. The findings of this study contribute to the development of edge computing traffic operation system accounting safety.
-
ItemNo Preview AvailableAbnormality Detection in Urban Traffic Data: A Review(Australasian Transport Research Forum, 2022)Anomalous data is called to a data sample or a sequence of data that significantly differs from the others. Accurately and on-time detection of anomalies (abnormalities) is crucial for system managers since it may convey important information to them. Anomaly detection is widely investigated in different research areas as well as transportation and traffic field. In this paper, we review the literature of anomaly detection in urban traffic networks to find the most recent state-of-the-art methodologies in this field. A search method is used in this paper to find the most relevant research papers, and they are studied and analyzed regarding anomaly type, data type, and methodology. Different types of anomalies, data collectors, spatiotemporal scopes, and detection methods in the literature are categorized and investigated in this work. Finally, a summary and conclusion section is provided in this work to show the possible future research directions. Based on the findings, accidents and city-wide events like festivals or concerts are mostly detected in previous works as anomalies using loop detector (LD), and trajectory data (GPS data). Moreover, supervised methods are mostly employed for accident detection aims, but papers using unsupervised approaches detect city-wide events with GPS data.
-
ItemNo Preview AvailableThe impact of high-occupancy lanes on the uptake of on-demand ridesplitting services(Australasian Transport Research Forum, 2022)New mobility alternatives such as on-demand ridesplitting services can bring large benefits to cities by increasing automobile occupancy and reducing congestion, pollution, and space allocated to parking. However, the current adoption of on-demand ridesplitting services is still limited and transport demand management (TDM) strategies may be necessary to increase such uptake. This work uses the agent-based simulation tool MATSim to simulate 10% of the population in the Greater Melbourne Area and investigate the effectiveness of dedicated ridesplitting lanes (DRL) on the uptake of on-demand ridesplitting services and overall transport network efficiency. Results suggest that the tested DRL configurations are effective in increasing the uptake of such services. We observe a significant increase in vehicle occupancy and a reduction in vehicle kilometres travelled, which indicate that this is a promising policy. However, in regard to average travel time, DRL scenarios benefit people with trips within the on-demand ridesplitting service’s area while deteriorating average travel time for people whose trips’ origins and/or destinations are outside the service’s area. Future simulations should incorporate multi-modal travel and include transport hubs that facilitate inter-modal transfers to test whether the observed benefits can be expanded to the areas outside the ridesplitting service area.
-
ItemNo Preview AvailableDeveloping a shopping destination choice model for people living in Metropolitan Melbourne(Australasian Transport Research Forum, 2022)One key factor to assist effective infrastructure and urban planning is to better understand the travel demands and the impact from various attributes on travellers’ decisions for destination. This study develops a destination choice model for shopping trips in Melbourne. The model predicts travellers’ choices for their shopping destinations from different areas in Melbourne based on demographic and trip related characteristics of individuals recorded in the Victorian Integrated Survey of Travel and Activity (VISTA) between 2012 and 2016. One of the main challenges overcome in this research is the choice set formation. ASGS Statistical Area 1 (SA1) has been adopted as zones for categorising origins and destination locations. For each observed trip from the Victoria travel survey, nine alternative destinations within 10km travel distance from the origin are randomly generated to form the choice set together with the observed destination. The final model is calibrated and to have up to 67% accuracy in the validation process and obtain a goodness of fit of 0.65, demonstrating promising performance. The final model indicates that three main attributes, travel distance, the number of shops and supermarkets in the destination precinct has the most significant impact on the choice of shopping destination in Melbourne. In addition, trip-specific attributes such as the number of stops made during a trip, the intended dwelling time at the destination and whether the travel is made during peak time (10am – 3pm) have varying levels of impact to destination choice on top of the three main attributes.