Mass Transit Railway Research Articles

Predicting metro incident duration using structured data and unstructured text logs

Predicting metro incident duration is crucial for passengers and transit operators to choose appropriate response strategies. Most existing research focuses on structured data, the rich information embedded within unstructured incident logs is often neglected. This paper incorporates a probabilistic topic model tailored for short texts, the biterm topic model, into the generic incident duration prediction models. By capturing text co-occurrence patterns through Bayesian inference, the biterm topic model extracts hidden topics from incident narratives, and each topic serves as a condensed summary of detailed incident causes and countermeasures. These extracted topics are then combined with structured information to serve as predictors. We validated our model using five years of incident data from the Hong Kong Mass Transit Railway across two scenarios: with all incident information available and information revealed over time . The results demonstrate that our method significantly improves the prediction accuracy, particularly for incidents lasting longer than 30-minute.

Self-calibrated transit service monitoring using automated collected data

This paper proposes a self-calibrated transit service monitoring framework that aims to obtain the performance of a transit system using automated collected data. We first introduce an event-based transit simulation model, which allows the detailed simulation of passenger travel behavior in a transit system, including boarding, alighting, and transfer walking. To estimate passenger path choices, we assume the path choices can be modeled using a C-logit model, and propose a simulation-based optimization model to estimate the path choice parameters based on automated fare collection and automated vehicle location data. The path choices can be estimated on a daily basis, which enables the simulation model to adapt to dynamic passenger behavior changes, and output more accurate network performance indicators for regular service monitoring such as train load, passenger travel time, and crowding at platforms. The proposed system eliminates the need for conventional monitoring equipment such as cameras at platforms and scaling/weighing systems on trains. The Hong Kong Mass Transit Railway (MTR) system is used as the case study. Results show that the model can well estimate the path choice behavior of passengers in the system. The output passenger exit flows are closer to the actual one compared to the two benchmark models (shortest path and uniform path choice).

Relative contributions of ambient air and internal sources to multiple air pollutants in public transportation modes

Commuters are often exposed to relatively high air pollutant concentrations in public transport microenvironments (TMEs) because of their proximity to emission sources. Previous studies have mainly focused on assessing the concentrations of air pollutants in TMEs, but few studies have distinguished between the contributions of ambient air and internal sources to the exposure of commuters to air pollutants. The main objective of this study was to quantify the contributions of ambient air and internal sources to the measured particulate matter and gaseous pollutant concentrations in selected TMEs in Hong Kong, a high-rise, high-density city in Asia. A sampling campaign was conducted to measure air pollutant concentrations in TMEs in Hong Kong in July and November 2018 using portable air quality monitors. We measured the concentrations of each pollutant in different TMEs and quantified the infiltration of particulate matter into these TMEs. The double-decker bus had the lowest particulate matter concentrations (mean PM1, PM2.5, and PM10 concentrations of 5.1, 9.5, and 13 μg/m3, respectively), but higher concentrations of CO (0.9 ppm), NO (422 ppb), and NO2 (100 ppb). For all the TMEs, about half of the PM2.5 were PM1 particles. The Mass Transit Railway (MTR) subway system had a PM2.5/PM10 ratio of about 0.90, whereas the PM2.5/PM10 ratio was about 0.60–0.70 for the other TMEs. The MTR had infiltration factor estimates <0.4 for particulate matter, lower than those of the double-decker bus and minibus. The MTR had the highest contribution from internal sources (mean PM1, PM2.5, and PM10 concentrations of 4.6, 13.4, and 15.8 μg/m3, respectively). This study will help citizens to plan commuting routes to reduce their exposure to air pollution and help policy-makers to prioritize effective exposure reduction strategies.

Developing a survival probability-based Lagrangian model for predicting ultraviolet disinfection of bioaerosols in enclosed environments

The transmission of airborne infectious diseases in enclosed environments has become a significant public health issue. Far-ultraviolet C (UVC) lights can be used safely and effectively to disinfect bioaerosols and reduce the risk of airborne infectious disease transmission. To ensure proper use of far-UVC systems, an improved Lagrangian method for predicting the dispersion, deposition, and far-UVC disinfection of bioaerosols was proposed. First, an improved Lagrangian model considering the probability of bioaerosol survival was developed for predicting the UVC disinfection of airborne bioaerosols in enclosed environments. The UVC disinfection of surfaces with deposited bioaerosols was implemented into the deposition module of the Lagrangian model by introducing the concept of equivalent deposition number. Experimental data on far-UVC disinfection of E. Coli in the air and on the surfaces in a ventilated chamber were used to validate the proposed model. The validation results show that the proposed model can predict the bioaerosol dispersion, deposition, and far-UVC disinfection reasonably well. The validated model was then employed to calculate the far-UVC disinfection of airborne and surface SARS-CoV-2 bioaerosols in a Mass Transit Railway (MTR) train compartment in Hong Kong. According to the calculation results, the far-UVC lamps are effective in reducing the passengers’ exposure to SARS-CoV-2 bioaerosols.

Vulnerability assessment of urban rail transit in face of disruptions: A framework and some lessons from Hong Kong

Transport systems are confronting challenges under disruptive events. In transit-oriented cities, the heavy reliance on urban railways makes the transport system ever more vulnerable. Extant literature has paid much attention to urban rail transit vulnerability and resilience against natural disasters. Yet, less focus has been put on disturbances caused by people. In this study, by integrating the on- and off-road situations, we assess the urban rail transit vulnerability under the Anti-extradition Law Amendment Bill (Anti-ELAB) movement in Hong Kong in 2019, during which the Mass Transit Railway (MTR) services were severely disrupted. Our results show that very highly vulnerable stations are spatially concentrated in traditional urban core areas. It is also observed that roads around some new town stations became more congested during service disruption. When disruptive events occurred during non-peak hours, the level of congestion in more than half of stations’ ambient roads did not show significant changes. This study sets forth a rail transit vulnerability assessment matrix and suggests that future assessments of transport vulnerability should take a more holistic approach and consider both on- and off-road traffic situations. The findings also provide directions for building resilient cities in the future.

Data-Driven Real-Time Denied Boarding Prediction in Urban Railway Systems

Providing real-time crowding information in urban railways would enable informed travel decisions and encourage cooperative behavior of passengers, as well as improve operating efficiency and safety. However, the problem of real-time crowding prediction is not trivial because of the unavailability of ground-truth crowding data, particularly for the direct impact of crowding on passengers (e.g., denied boarding on platforms). This paper proposes a data-driven method for real-time denied boarding prediction in urban railway systems using automated fare collection (AFC) and automated vehicle location (AVL) data. It predicts the denied boarding probability distribution or its derived metrics as a function of explanatory variables, including demand, operations, and incident-related factors. The method is validated through a case study covering 18 months on Hong Kong Mass Transit Railways. The results highlight the model’s accurate and robust performance in predicting denied boarding on platforms using purely AFC and AVL data (e.g., an average of 6%–7% error) under both recurrent and non-recurrent situations, in which transfer demand-related factors contribute most to the prediction. The prediction model outputs can support proactive operations control and management, as well as provision of customer information without relying on expensive and time-consuming data collection efforts. Customer information, for example, can include the average number of trains to wait before boarding or the average waiting time on the platform.

A Smartphone-Enabled IoT System for Vibration and Noise Monitoring of Rail Transit

With the development of the urban rail transit system, the ride comfort of trains in operation has been emphasized by passengers and railroad operators alike. The monitoring of interior vibration and noise levels of trains provides an efficient way to guarantee ride comfort. Traditional monitoring equipment has numerous disadvantages, such as high cost, poor mobility, and lack of real-time evaluation functions. To solve these problems, this study explored the utilization of smartphones to measure vibration and noise and diagnose human comfort levels. A smartphone application has been developed based on the Internet of Things (IoT) concept. The smartphone device not only realizes data sensing, visualization, and transmission, but also fulfills edge-computing functions to process sensing data and evaluate vibration and noise impact in real time. The design principle and major components of the developed application are elaborated. In situ experiments were conducted on the Hong Kong Mass Transit Railway, which provides valuable experiences for applying IoT monitoring technology in the rail transit systems and enables the realization of citizen sensing in the future.

Integrated optimization of bus bridging service design and passenger assignment in response to urban rail transit disruptions

As the urban rail transit (URT) system plays an increasingly important role in supporting large cities’ mobility around the world, service disruptions have become more prevalent, potentially resulting in severe economic losses and passenger safety issues. It is imperative to investigate effective response strategies to mitigate the effects of such disruptions. In response to URT service disruptions, this paper systematically investigates the bus bridging service design (BBSD) problem, which concerns the integration of bus bridging route design, frequency determination, and passenger assignment in the integrated URT and bus network. The problem is formulated as a path-based integer linear programming (ILP) model with the goal of simultaneously minimizing operator-oriented and passenger-oriented costs. A column generation-based approach is proposed to solve this model efficiently, allowing nonintuitive bus routes to be freely generated on the network dynamically. Our method has been tested with two different case studies based on real data from the Hong Kong Mass Transit Railway (MTR). Experiments demonstrate that our proposed approach can assist public transit (PT) operators in developing efficient emergency response plans for various potential disruption situations in advance. Even in the face of unexpected disruptions that necessitate a quick response, our approach can generate high-quality solutions in a matter of minutes.

Changes on local travel behaviors under travel reduction-related interventions during COVID-19 pandemic: a case study in Hong Kong

The emerging Omicron variant poses a serious threat to human health. Public transports play a critical role in infection spread. Based on the data of nearly 4 billion smartcard uses, between January 1, 2019 and January 31, 2021 from the Mass Transit Railway Corporation of Hong Kong, we analyzed the subway travel behavior of different population groups (adults, children, students and senior citizens) due to the COVID-19 pandemic and human travel behavior under different interventions (e.g. work suspension, school closure). Due to the pandemic, the number of MTR passengers (the daily number of passengers in close proximity in subway carriages) decreased by 37.4% (40.8%) for adults, 80.3% (78.5%) for children, 71.6% (71.6%) for students, and 33.5% (36.1%) for senior citizens. Due to work from home (school suspension), the number of contacted adults (students/children) in the same carriage during the rush hours decreased by 39.6% (38.6%/43.2%). If all workers, students, and children were encouraged to commute avoiding rush hours, the possible repeated contacts during rush hour of adults, children and students decreased by 73.3%, 77.9% and 79.5%, respectively. Since adults accounted for 87.3% of the total number of subway passengers during the pandemic, work from home and staggered shift pattern of workers can reduce the infection risk effectively. Our objective is to find the changes of local travel behavior due to the pandemic. From the perspective of public transports, the results provide a scientific support for COVID-19 prevention and control in cities.

Transition from fossil fuel propelled transport to electrified mass transit railway system - Experience from Hong Kong

Shifting from fossil fuel propelled vehicle to electrified transport becomes a growing trend and forms part of an energy and transportation policy in many modern cities. Due to improper design, management and operation, many mass transit companies are operating under a tough financial condition. This paper analyzes and discusses the strategy and policy for successful transition from fossil fuel dominant transportation to an electrified mass transit railway system, taking Hong Kong as a study case. The Hong Kong Mass Transit Railway (MTR) system has 98 stations along 10 domestic rail lines of 266.3 km route length, with an average daily passenger number of 4.189 million in this highly dense city with a population of 7.413 million. The Hong Kong MTR system shares 48% of the franchised transport boarding with the expense of only 4.9% of the annual energy consumption in the entire transport sector. In addition to the renowned business model: Rail plus Property model which has been quoted in many literatures, this study analyzes and discusses the design, operation, management, feeder bus service, and the government's policy and strategies on fossil fuel driven vehicle control. Moreover, suggestion on enhancing the existing contingency plan of the MTR operation is proposed.

Optimal congestion control strategies for near-capacity urban metros: Informing intervention via fundamental diagrams

Congestion; operational delays due to a vicious circle of passenger congestion and train-queuing; is an escalating problem for metro systems because it has negative consequences from passenger discomfort to eventual mode shifts. Congestion arises due to large volumes of passenger boardings and alightings at bottleneck stations, which may lead to increased stopping times at stations and consequent queuing of trains upstream. Such queuing further reduces line throughput and implies an even greater accumulation of passengers at stations. Alleviating congestion requires control strategies such as regulating the inflow of passengers entering bottleneck stations. The availability of large-scale smartcard and train movement data from day-to-day operations facilitates the development of models that can inform such strategies in a data-driven way. In this paper, we propose to model station-level passenger congestion via empirical passenger boarding–alightings and train flow relationships, henceforth, fundamental diagrams (FDs). We emphasise that estimating FDs using station-level data is empirically challenging due to confounding biases arising from the interdependence of operations at different stations, which obscures the true sources of congestion in the network. We thus adopt a causal statistical modelling approach to produce FDs that are robust to confounding and as such suitable to inform control strategies. The closest antecedent to the proposed model is the FD for road sections, which informs traffic management strategies, for instance, via locating the optimum operation point. Our analysis of data from the Mass Transit Railway, Hong Kong indicates the existence of concave FDs at identified bottleneck stations, and an associated critical level of boarding–alightings above which congestion sets in unless there is an intervention.

Sparse and Robust Multivariate Functional Principal Component Analysis for Passenger Flow Pattern Discovery in Metro Systems

Modern metro systems in big cities have accumulated large amounts of passenger transit transaction data via the deployment of automatic fare collection (AFC) devices, which could facilitate a thorough analysis of passenger flow dynamics. To discover the underlying passenger flow patterns over all stations in an entire metro system, this paper proposes a multivariate functional principal component analysis (MFPCA) method. The functional integrity of a single daily passenger flow profile at each station is explicitly utilized, and the complex correlation among a multitude of daily passenger flow profiles from all stations is comprehensively modeled. Moreover, to simultaneously improve the interpretability of eigenvectors and mitigate the sensitivity to outliers, the MFPCA is formulated as a minimization problem with both a sparsity and a robustness penalty term. A computationally efficient algorithm is developed accordingly to obtain the eigenvectors. The superiority of our proposed sparse and robust MFPCA (SRMFPCA) is validated using a Hong Kong Mass Transit Railway (MTR) dataset. The derived sparse and smooth eigenvectors can be well interpreted as empirically meaningful passenger flow patterns. The results of our method can be further used as solid foundations for station clustering, correlation analysis and outlier identification.

Application of artificial intelligence (AI) control system on chiller plant at MTR station

Chillers account for up to 40% of total station energy consumption in the Hong Kong Mass Transit Railway (MTR) system. As part of green railway initiatives, a site trial was conducted to apply a fully automated AI system to control a chiller plant in order to optimise energy performance in real time while maintaining a level of passenger comfort that suits each station’s environment. Through the predictive power of the AI system, the plant power’s consumption and cooling demands can be forecasted based on actual chiller, station, and weather conditions, all of which vary over time. The optimal operational settings can then be determined using an optimisation model for real-time chiller plant control, including staging, sequencing, chilled water supply temperature set-point, etc. This paper presents the formulation of an AI system using data-driven machine learning models and numerical optimisation, and the comparison of the actual energy performance of the proposed system against rule-based control optimisation in a conventional building management system (BMS) through the site trial. The results revealed the proposed AI system achieves better energy efficiency with annual energy savings of approximately 8.7%.

Nonparametric Passenger Flow Monitoring using A Minimum Distance Criterion

Monitoring real-time passenger flow in urban rapid transit systems is very important to maintain social stability and prevent unexpected group events and system failure. To monitor passenger flow, data are collected by sensors deployed in important stations and many existing control charts can be applied. However, because of unknown complex distributions and the requirement to detect shifts of all ranges effectively, conventional methods may perform poorly. Nevertheless, while there are certain charting schemes that truncate the log-likelihood ratio (LLR) function to detect large shifts more quickly, they can cause massive loss of information by truncation, and can only handle particular distributions, leading to unstable online monitoring. In this article, we propose a nonparametric CUSUM charting scheme to monitor passenger flow dynamically. We propose a novel minimum distance criterion is proposed to minimize the functional distance between the objective function and the original LLR function while maintaining its monotonically increasing property. By integrating this concept with kernel density estimation, our proposed chart does not require any parametric process distribution, can be constructed easily in any situation, and is sensitive to shifts of all sizes. Theoretical analysis, simulations and a real application to monitoring passenger flow in the mass transit railway (MTR) in Hong Kong show that our method performs well in various cases.

Does rail transit development gentrify neighborhoods? Evidence from Hong Kong

Transit-induced gentrification is a popular topic of research and a pressing concern worldwide. However, little attention is paid to the connection between rail transit development and the movement of low-income residents. To investigate this important issue, this paper studies the development of the Mass Transit Railway (MTR) in Hong Kong between 2001 and 2006, which added three new rail transit lines covering a total length of 65 km. By using census statistics at the district council constituency area (DCCA) level in a difference-in-differences approach, we find that MTR development in Hong Kong between 2002 and 2004 increased the number of advanced degree holders (Bachelor’s degree and above) by 10% in the New Territories and 7.81% in the new towns and reduced the number of low-income households by 5.85% in the New Territories and 7.53% in the new towns. To confirm that these socioeconomic changes were driven by internal migration, we further use household-level data from the 5% population censuses in 2001 and 2006 in a multinomial logistic regression. The results indicate that MTR development between 2002 and 2004 was likely to induce high-educated people to move into and low-income households to move out of MTR-served areas. The results show that the New Territories and the new towns in Hong Kong underwent gentrification between 2001 and 2006. We therefore suggest that the Hong Kong government negotiate with MTR Corporation to allocate resources to develop properties for low-income families.

A Survey on Formal Specification and Verification of Smart Mass Transit Railway Interlocking System

Nowadays interest in Smart Mass Transit Rail has grown-up to a large extent in a metropolitan area as the need for urban mobility has increased steadily. The reliability of software being used in such mass transit rail is crucial for us, specifically when software crashes may lead to catastrophic loss of human life and assets. For example, when we travel by metro it is essential for us that the interlocking system software controlling the metros are accurate so collisions and derailment are prevented. The reliability and safety of such interlocking systems are made on the precise functional requirements specification and verification respectively. Therefore, the precise functional requirements specification and verification of such interlocking systems represent a challenge in an active research area, so in this paper, we survey various articles in this field and discuss their consequences.

What moves Hong Kong's train ridership?

Hong Kong is a densely populated international metropolis with ∼7.5 million residents living in a small geographic area of ∼1,100 km2. Like some major cities around the world, it has a vast public transportation system that moves ∼12.4 million passengers per day, ∼42% of which is attributable to the Mass Transit Railway's (MTR's) extensive network interconnecting widely dispersed stations across Hong Kong. MTR's ridership substantially declined in 2019 because of social unrest and related system service suspension (SSS). This decline was further exacerbated by Covid-19's outbreak in 2020. Using a hand-collected sample of monthly data for January 2000–June 2020, we estimate a log-linear demand regression to find that MTR's ridership is price and income inelastic, varies seasonally, and exhibits a trend of gradual growth. Further, Covid-19, social unrest and SSS significantly reduce MTR's ridership. Finally, the estimates for passenger welfare losses due to SSS and social unrest are large, revealing the potential gains from improving MTR's service reliability and resolving the root causes of social unrest.

Feasibility analysis of GNSS-based ionospheric observation on a fast-moving train platform (GIFT)

The ionospheric effect plays a crucial role in the radio communications. For ionospheric observing and monitoring, the Global Navigation Satellite System (GNSS) has been widely utilized. The ionospheric condition can be characterized by the Total Electron Contents (TEC) and TEC Rate (TECR) calculated from the GNSS measurements. Currently, GNSS-based ionospheric observing and monitoring largely depend on a global fiducial network of GNSS receivers such as the International GNSS Service (IGS) network. We propose a new approach to observe the ionosphere by deploying a GNSS receiver on a Hong Kong Mass Transit Railway (MTR) train. We assessed the TECR derived from the MTR-based GNSS receiver by comparing it with the TECR derived from a static GNSS receiver. The results show that the Root-Mean-Squares (RMS) errors of the TECR derived from the MTR-based GNSS receiver is consistently approximately 23% higher than that derived from the static GNSS receiver. Despite the increased error, the findings suggest that the GNSS observation on a fast-moving platform is a feasible approach to observe the ionosphere over a large region in a rapid and cost-effective way.

Unveiling the rich-club phenomenon in urban mobility networks through the spatiotemporal characteristics of passenger flow

With increasing passenger flows and rapid urbanization, the Mass Transit Railway (MTR) systems in Hong Kong have entered into an era of complex networking operation, which enables people to travel across 18 local districts and enhance sustainable transport accessibility. The urban mobility flow patterns typically encoded as origin–destination (OD) matrices inspire the constitution of the temporal urban mobility networks (TUMNs), which are represented as the directed weighted networks. The dynamics of the TUMNs have significant implications for transportation systems, urban traffic planning, and mobility pattern exploration. This paper provides a thorough analysis of the rich-club phenomenon of TUMNs in Hong Kong during different time periods. Additionally, the space–time characteristics of passenger flow are analyzed and based on which we build the directed weighted TUMNs where the nodes are the stations, and the weighted links represent the number of trips between nodes. By using the outgoing-strength and incoming-strength global rich-club coefficient, local rich-club coefficient, community detection, and assortativity coefficient, we quantitatively identify the rich-club phenomenon in the TUMNs during both the peak hours (8:00–9:00 and 18:00–19:00) and non-peak hours (10:00–11:00, 15:00–16:00, and 20:00–21:00). The findings uncover the rich-club phenomenon in the TUMNs, and the distribution of the rich-club members evolving over time. The findings of this research pave the way towards a richer spectrum of complex real-world networks and are useful for transport infrastructure planning.

Performance-based regulation of train service delays: Hong Kong

ABSTRACT This paper studies performance-based regulation of train service delays in Hong Kong. Motivated by the criticism that the Mass Transit Railway (MTR) offers fare rebates deemed too small for long service delays, our logit analysis of a stated choice experiment’s data collected from 1,559 MTR users in Q4:2020 yields the following findings: (1) the disaggregate willingness to pay (WTP) estimates that ignore passenger attributes are HK$21 to HK$37 for a 15- to 60-minute train service delay announced to occur in a 1-hour period; (2) increasing the period by one hour raises these estimates by ~HK$16; (3) accounting for passenger attributes magnifies these estimates by ~18%; and (4) the aggregate WTP estimates are 1.3 to 4.9 times MTR’s existing rebates for rush-hour delays and 0.5 to 1.9 times for non-rush-hour delays. Hence, aligning MTR’s rebates with the aggregate WTP estimates likely incentivizes MTR to reduce long service delays.