Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sohani Liyanage has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Real-World Emissions and Range Performance of Passenger Vehicles in Australia

    by Sreedhar Harikumar Kartha, Hussein Dia, Sohani Liyanage

    Laboratory test results for vehicle emissions, fuel economy, and driving range often fail to reflect real-world performance, undermining the effectiveness of sustainability policies and consumer guidance. This study provides the first integrated national assessment of real-world emissions and range outcomes for passenger vehicles in Australia. Using Portable Emissions Measurement Systems (PEMS) data from 114 petrol, diesel, hybrid, and battery-electric vehicles (BEVs) tested by the Australian Automobile Association (AAA), the analysis compares laboratory-certified values against on-road results and benchmarks them with international datasets from Europe and China. Real-world CO2 emissions were, on average, 6.9% higher than laboratory ratings for petrol vehicles and 3.2% higher for diesel vehicles. Many diesel models exceeded Euro 6 NOx limits by several multiples, while hybrids exhibited inconsistent CO2 reductions under urban conditions. BEVs also displayed measurable divergence: real-world energy consumption was 1–20% higher than laboratory ratings, resulting in an average 16% reduction in effective driving range relative to WLTP values. These outcomes reveal a consistent tendency toward overstated laboratory performance across powertrains, highlighting systemic shortcomings in certification test cycles. The findings have direct implications for greenhouse gas mitigation, urban air quality, and consumer energy efficiency and support Australia’s active transition to WLTP and Euro 6 standards, institutionalisation of real-world testing, and inclusion of verified real-world energy use and range data in consumer labelling to enhance transparency and policy effectiveness.


  • Open Access

    AI-Powered Evaluation of On-Demand Public Transport: A Hybrid Simulation Approach

    by Sohani Liyanage, Hussein Dia, Gordon Duncan

    On-demand public transport systems are increasingly adopted to improve service flexibility, reduce operating costs, and meet emerging mobility needs. Evaluating their performance under realistic demand and operational conditions, however, remains a complex challenge. This study presents a hybrid simulation framework that integrates deep learning-based demand forecasting, behavioural survey data, and agent-based simulation to assess system performance. A BiLSTM neural network trained on real-world smartcard data forecasts short-term passenger demand, which is embedded into an agent-based model simulating vehicle dispatch, routing, and passenger interactions. The framework is applied to a case study in Melbourne, Australia, comparing a baseline fixed-route service with two on-demand scenarios. Results show that the most flexible scenario reduces the average passenger trip time by 32%, decreases the average wait time by 34%, increases vehicle occupancy from 12.1 to 18.6 passengers per vehicle, lowers emissions per passenger trip by 72%, and cuts the service cost per trip from AUD 6.82 to AUD 4.73. These findings demonstrate the potential of hybrid on-demand services to improve operational efficiency, passenger experience, and environmental outcomes. The study presents a novel, integrated methodology for scenario-based evaluation of on-demand public transportation using real-world transportation data.


  • Top Cited Papers
    Open Access

    Machine Learning Traffic Flow Prediction Models for Smart and Sustainable Traffic Management

    by Rusul Abduljabbar, Hussein Dia, Sohani Liyanage

    Sustainable traffic management relies on accurate traffic flow prediction to reduce congestion, fuel consumption, and emissions and minimise the external environmental impacts of traffic operations. This study contributes to this objective by developing and evaluating advanced machine learning models that leverage multisource data to predict traffic patterns more effectively, allowing for the deployment of proactive measures to prevent or reduce traffic congestion and idling times, leading to enhanced eco-friendly mobility. Specifically, this paper evaluates the impact of multisource sensor inputs and spatial detector interactions on machine learning-based traffic flow prediction. Using a dataset of 839,377 observations from 14 detector stations along Melbourne’s Eastern Freeway, Bidirectional Long Short-Term Memory (BiLSTM) models were developed to assess predictive accuracy under different input configurations. The results demonstrated that incorporating speed and occupancy inputs alongside traffic flow improves prediction accuracy by up to 16% across all detector stations. This study also investigated the role of spatial flow input interactions from upstream and downstream detectors in enhancing prediction performance. The findings confirm that including neighbouring detectors improves prediction accuracy, increasing performance from 96% to 98% for eastbound and westbound directions. These findings highlight the benefits of optimised sensor deployment, data integration, and advanced machine-learning techniques for smart and eco-friendly traffic systems. Additionally, this study provides a foundation for data-driven, adaptive traffic management strategies that contribute to sustainable road network planning, reducing vehicle idling, fuel consumption, and emissions while enhancing urban mobility and supporting sustainability goals. Furthermore, the proposed framework aligns with key United Nations Sustainable Development Goals (SDGs), particularly those promoting sustainable cities, resilient infrastructure, and climate-responsive planning.


  • Open Access

    On-Demand Technologies for Public Transport: Insights From a Melbourne Survey

    by Sohani Liyanage, Hussein Dia

    The integration of on-demand technologies in urban mobility requires a comprehensive understanding of user acceptance and willingness to pay for innovative modes like on-demand public transport designed to enhance conventional services. This study presents findings from a survey conducted in Melbourne, highlighting passenger behaviours, preferences, and attitudes towards the use of on-demand transport technologies as a sustainable alternative to conventional bus services. Data from 510 diverse participants revealed a strong preference for private vehicles, mainly for convenience and flexibility. However, concerns regarding waiting times, crowding, and reliability in public transport highlighted the need for service improvements. The survey included hypothetical scenarios where respondents evaluated on-demand transport options with varying factors like waiting time, travel cost, and journey duration. Using binary logistic regression and neural networks (NN), the study analysed preferences for the proposed hypothetical on-demand transport scenarios, revealing that while travel cost negatively impacts mode choice, reduced waiting times positively influence it. The binary logistic model showed classification accuracies between 64% and 72%, while the NN models achieved a high prediction accuracy, reaching approximately 91%. The results indicate that 67% would switch to on-demand transport if it offered reliability, convenience, reduced crowding, and fair pricing. Additionally, 53% were willing to pay a 25% premium for shorter walking and waiting times, with 69% identifying service reliability as the key factor influencing their transport decisions. These insights are essential for developing transport technology frameworks that incorporate on-demand technologies within existing public transport systems, thus advancing sustainable and resilient urban mobility solutions.


  • Open Access

    Machine Learning Models for Traffic Prediction on Arterial Roads Using Traffic Features and Weather Information

    by Rusul Abduljabbar, Hussein Dia, Sohani Liyanage

    This study addresses the challenges of predicting traffic flow on arterial roads where dynamic behaviours such as passenger pick-ups, vehicle turns, and parking interruptions complicate forecasting. The research develops and evaluates unidirectional and bidirectional Long Short-Term Memory (LSTM) models using a dataset of 70,072 observations collected over 12 months from Hoddle Street in Melbourne, Australia. The models were trained to predict traffic speeds and volumes up to 60 min ahead. The results indicated that the BiLSTM model significantly outperformed unidirectional LSTM, achieving over 99% accuracy for speed predictions and over 96% for volume predictions. The research also tested the impacts of incorporating weather variables such as rainfall, temperature, humidity, and wind speed on model performance, which was found to provide small improvements. Traffic flow prediction accuracy increased from 97.5% to 97.6% for 30-min horizons, and from 96.9% to 97.6% for 60-min horizons. Although the inclusion of weather data only marginally enhanced prediction performance, its inclusion has practical implications for public awareness of travel impacts under severe weather. The findings in this study highlight the effectiveness of deep learning techniques for traffic forecasting on arterial roads, establishing a foundation for future research in this area.


  • Open Access

    Evaluation of the Impacts of On-Demand Bus Services Using Traffic Simulation

    by Sohani Liyanage, Hussein Dia, Gordon Duncan, Rusul Abduljabbar

    This paper uses smart card data from Melbourne’s public transport network to model and evaluate the impacts of a flexible on-demand transport system. On-demand transport is an emerging mode of urban passenger transport that relies on meeting passenger demand for travel using dynamic and flexible scheduling using shared vehicles. Initially, a simulation model was developed to replicate existing fixed-schedule bus performance and was then extended to incorporate on-demand transport services within the same network. The simulation results were used to undertake a comparative analysis which included reliability, service quality, operational efficiency, network-wide effectiveness, and environmental impacts. The results showed that on-demand buses reduced average passenger trip time by 30%, increased vehicle occupancy rates from 8% to over 50%, and reduced emissions per passenger by over 70% on an average weekday compared to fixed-schedule buses. This study also offers insights for successful on-demand transport implementation, promoting urban sustainability. It also outlines future research directions, particularly the need for accurate short-term passenger demand prediction to improve service provision and passenger experience.


  • Neural network approaches for forecasting short-term on-road public transport passenger demands

    by Sohani Liyanage, Hussein Dia, Rusul Abduljabbar, Pei-Wei Tsai

    On-road public transport plays an essential role in urban mobility and in meeting people's needs for travel. Service reliability and punctuality are important factors in encouraging more people to use these forms of transport and improving passenger user experiences in a modern world where passengers expect higher levels of service availability and reliability. On-road public transport operators can improve services when they have access to reliable estimates of passenger demands, which helps them to adjust timetables for each service route efficiently based on average passenger demands at public transport stops or on a service route at a specific time. This chapter presents an overview of the extensive research performed in the recent decade on AI-based passenger demand forecasting for public transport systems. The chapter also provides insights into the potential of AI-based prediction, and how it has been shown to achieve high forecasting accuracy exceeding 90% using real-world datasets.


  • Open Access

    AI-based neural network models for bus passenger demand forecasting using smart card data

    by Sohani Liyanage, Rusul Abduljabbar, Hussein Dia, Pei-Wei Tsai


  • Performance Evaluation of Station-Based Autonomous On-Demand Car-Sharing Systems

    by Farid Javanshour, Hussein Dia, Gordon Duncan, Rusul Abduljabbar, Sohani Liyanage

    Autonomous Mobility-on-Demand (AMoD) systems hold potential promise for addressing urban mobility challenges. Their key principle is to utilize fleets of shared self-driving vehicles to respond to customer demand on flexible routes in real-time. This research investigates station-based AMoD car-sharing systems and uses scenario analyses to identify plausible future paths for their deployment. A traffic simulation model which implements real-time rebalancing of idle vehicles is developed to evaluate their performance under uncertain travel demands. Unlike other literature which assumed homogeneous demand and resulted in low increases in vehicle kilometers travelled (VKT), this study relied on realistic heterogeneous demand and showed a significant increase in VKT. A case study for Melbourne demonstrated the impacts and showed that while AMoD can meet the demand for travel using only 16% of the current vehicle fleet, they would produce 77% increase in VKT. This would significantly increase congestion in any real-world scenario and goes against the hype of AMoD being the answer to congestion problems.