Ebook: Emerging Cutting-Edge Developments in Intelligent Traffic and Transportation Systems

The International Conference on Intelligent Traffic and Transportation (ICITT) is a major event for academics, researchers, and industrialists who are engaged in Intelligent Traffic and Transportation research. Held annually since 2010, the conference is renowned as a friendly and inclusive platform that brings together a broad community of researchers who share a common goal: developing and managing the engineering and technologies relevant to the revolution in transportation systems, and the operations key to sustaining the success of intelligent traffic and transportation industries. For over a decade, ICITT has been the main research conference organized worldwide which deals with intelligent traffic and transportation, and successfully brings together researchers, academics, and industrialists to share their knowledge, expertise and experience. Initiated as an national intelligent traffic and transportation workshop by the ICITT Consortium, a group of prominent university professors working in applied intelligent traffic and transportation research, it became an annual international conference in 2016.

The ICITT Consortium is an independent body established in 2010. Its main aim is to promote intelligent traffic and transportation engineering, technological education, training and research, and knowledge exchange and transfer. To achieve this, the conference chairman, consortium, and committee members maintain a close association with a number of international bodies concerned with the training and continuing development of professional engineers and technologists, while responding to appropriate consultations and discussion of documents, and other initiatives. The conference also plans major industrial visits to small and medium enterprises (SMEs) and large enterprises (LEs) in the industry to enhance the knowledge and experience of conference participants as regards the current and future industrial revolution (i.e., digital transformation, industry 4.0 and beyond, autonomous technology, industrial automation, etc.) and its impact on the intelligent traffic and transportation market.

In 2016, the ICITT conference was given the title International to reflect current trends in intelligent traffic and transportation engineering and technology and to promote the international exchange of research, engineering, and technology application experience. In previous years, the ICITT has taken place in the following countries:

ICITT 2016 China

ICITT 2017 France

ICITT 2018 Sweden

ICITT 2019 Netherlands

ICITT 2020 VR Online

ICITT 2021 VR Online

ICITT 2022 France

ICITT 2023 Spain

Traffic congestion is one of the biggest problems for transportation, since it carries high costs for cities and health risks due to its impact on air pollution. Therefore, several strategies have been proposed to improve mobility considering pollution reduction as a critical factor. To accomplish that, methods like forecasting and traffic light control systems are used for traffic management and for them to be more efficient, prior analysis of traffic data is needed. This paper is focused on that analysis for a Low Emission Zone located in Medellin, Colombia. The most relevant characteristics related to the temporal component and variables of the dataset are visualized and analyzed through statistical methods such as correlation and visualization approaches such as PCA and bar plots, also used for feature selection.

One of the biggest challenges in validating the electronic equipment of vehicles is finding suitable methods for virtual testing and simulating real-world scenarios as accurately as possible. Although computer simulations are safe and reproducible, there are significant simulation-to-reality gaps, making safety testing within simulations unreliable. Due to the lack of Precise sensor and traffic models, The data generated through simulation appears to be relatively realistic, still cannot replicate all the details of the real world. In this study, we propose to construct a secure and reliable assessment and validation platform by leveraging the combination of augmented reality technology and vehicle-in-the-loop simulation technique, which is called augmented reality-based proving ground vehicle-in-the-loop test platform. The method aims to combine real-world and virtual testing, making it easier and safer to test autonomous vehicles in critical scenarios while optimizing the validation process. Our proposed system offers an improved approach by combining simulated sensor data with real sensor data collect to generate augmented reality scenario data, which include AR based BUS sensor, AR based camera and AR based Lidar, providing more precise data support for the perception and decision-making processes of autonomous vehicles. In summary, the above-mentioned method provides a more comprehensive and accurate way of simulating scenarios, which can help improve the performance and safety of autonomous vehicles in the real world. Finally, we demonstrate the broader implications that such a simulation paradigm may have for autonomy, specifically showing how realistic sensor simulation can improve perception performance.

Given the rising attention towards the understanding of people’s mobility, this paper focuses on the study of next-place prediction models for mobility demand and path estimation. A systematic literature review was conducted to classify the existing methodologies from a quantitative and qualitative perspective. The findings highlight the availability of several models to study next-place prediction which varies according to the objectives of the study and on the data used, with no specific dominant approach. In conclusion, the study proposes a first attempt of classification, by developing a conceptual framework, that on the one hand explains the relationship between the models’ characteristics and on the other hand guides its selection according to the users’ needs.

Currently, the automotive industry has been adopting new validation methods for its processes during the development and test phases of new technologies. As part of this, the intention of reducing time and costs of physical tests, achieving the reduction of material and human risks. However, this leads to new challenges and problems such as human interaction of the tested technologies. One of these methods consists on the development and validation of new systems using computer simulations due to the high capacity of computer processors, being able to recreate virtually physical measurements of the environment. Nowadays it is used to predict the behavior, reliability and durability of materials, systems, com-ponents, etc. In other hand, the reproduction of realistic traffic in computer simulations remains a major challenge. In this paper we address the driver behavior modeling, a brief description of the use of road safety reports, in order to identify the different objective factors (weather, road condition, driving time…) and subjective factors (fatigue, stress...) that directly or indirectly affect the driver’s behavior. Then, integrate them to build a model with a modular and flexible architecture using a fuzzy logic approach, which will be able to reproduce the behavior of an average driver considering the influence of several factors. Thus, the model will allow to generate a more realistic and natural traffic in numerical simulations. The proposed model was validated in simulation using SCANeR Studio software and the results obtained showed the effectiveness of the proposed model to represent the influence of the characteristics of the driver profile.

Motorcycle crashes are a significant cause of injuries and deaths worldwide, with risky driving behaviors among motorcyclists being a significant contributing factor. This study aimed to identify the factors associated with risky driving behaviors among motorcyclists in an urban setting in Ecuador. Data were collected through observation of riders in the city of Loja, focusing on compliance with helmet usage, unfastened helmets, and distractions caused by cell phone usage while driving. The decision tree method was used to analyze the data. The results showed that wearing a helmet without fastening it and being distracted by the cell phone while driving were the most significant factors associated with unsafe driving behaviors among motorcyclists in Ecuador. Likewise, the riders of delivery motorcycles show more insecure behavior than those driving private motorcycles, especially when they are close to the economic center of the city. These findings can inform the development of road safety campaigns and the enactment of laws to reduce motorcycle accidents and improve rider safety.

Efficient automatic detection of incidents is a well-known problem in the field of transportation. Non-recurring incidents, such as traffic accidents, car breakdowns, and unusual congestion, can have a significant impact on journey times, safety, and the environment, leading to socio-economic consequences. To detect these traffic incidents, we propose a framework that leverages big data in transportation and data-driven Artificial Intelligence (AI)-based approaches. This paper presents the proposed methodology, conceptual and technical architecture in addition to the current implementation. Moreover, a comparison of data-driven approaches is presented, the findings from experiments to explore the task using real-world datasets are examined, while highlighting limitations of our work and identified challenges in the mobility sector and finally suggesting future directions.

Urban mobility congestion is a serious threat to the economic prosperity and way of life for individuals integrated into city infrastructures. Solutions for this problem require rethinking the time and spatial redesign of urban locations. In recent work, we developed a systemic and bi-disciplinary methodology to predict people’s willingness and flexibility to shift their working hours based on sociological criteria. This paper extends this work by developing a new approach to analyze traffic jams and proposing the best departure time for users to avoid traffic congestion and contribute to reducing urban mobility congestion. The developed methodology relies on three phases. First, using the Open Source Routing Machine (OSRM), the routes that an individual passes by to go to their workplace are generated. Second, the traffic of each route is analyzed according to different parameters (daytime, weekday, month, built environment, weather conditions, etc.) using Waze data; the time series model FB Prophet then predicts the congestion factor of different timings. Finally, the best (i.e., shortest spending travel time) departure time is extracted.

Road designs that prioritize vehicular flow over vulnerable users are common in Latin America. The present investigation evaluates the impact of the inclusion of a pedestrian bridge, in an urban intersection, on pedestrian mobility. Based on the field inspection and video camera recordings, the recurring behavior of pedestrians at crosswalks and desire lines was identified, lines that are often illegal crossings. In addition, the flow and pedestrian density on the observed routes were determined. It was observed that the pedestrian bridge is used by 24% of pedestrian users at the intersection, the rest of pedestrians use the two informal routes and the two formal ones, which indicates a poor urban infrastructure design. Therefore, a safe alternative design of the intersection was made based on the pedestrian desire lines. A proposal of crossing paths and two pedestrian traffic lights was made, which also include removing the current pedestrian bridge. Therefore, the impact of this bridge on pedestrian mobility was evaluated. The evaluation was carried out through a micro simulation with the VISSIM-VISWALK Software. Results indicate that the crossing travel time is reduced from 107.99 seconds with a pedestrian bridge, to 27.52 seconds without a pedestrian bridge. In addition, there is a reduction in the probability of collision, from 15 incidents of low risk and 2 of high risk, to 9 and 0.5, respectively. Therefore, in this type of urban intersection, the pedestrian bridge does not show any improvement to the pedestrian mobility; on the contrary, due to having a larger travel path which brings more travel time, a few users use this infrastructure without any safety benefit. And, with non-significant changes in the design and including traffic devices such as pedestrian traffic lights, there is a significant improvement in terms of mobility and safety.

This study proposes a vision-based high-speed localization estimation method for location based visual inspection of specific cracks in tunnels on Japanese expressways where global navigation satellite systems are not applicable. The method relies on recognizing lighting facilities installed in tunnels by using random sample consensus (RANSAC), enabling stable and accurate localization estimation in the horizontal direction. To correspond with traveling at high speed, single instruction/multiple data (SIMD) conversion realized 8 times faster than conventional image processing. The evaluation experimental results on expressway demonstrate that the proposed method achieves a maximum error of 31 mm in estimating lighting facilities position with an average error of 16 mm. The theoretical value derived from tunnel completion drawings has a maximum difference of 177 mm from the total value by this method, indicating that the results of accurate on-site measurements should be prioritized over completion drawings. In conclusion, the proposed method has considerable potential for practical application in tunnel inspection and maintenance.

Effective assignment of delivery task to multiple drivers is crucial for enabling smooth operation and maximizing end-user satisfaction. This paper focuses on batch assignment, i.e., the allocation of tasks among multiple drivers, starting from the same depot, while taking into account a variety of factors like load balancing delivery time, distance, workload balance, and overall effectiveness. A Kmeans Constrained technique is proposed in order to address the problem of minimizing the execution time during the assignment process and simultaneously taking into account constraints like the total number of orders, load capacity, and the number of drivers.

Predicting the intentions of other vehicles in traffic is a frequently addressed challenge in autonomous driving. Due to the complexity and diversity of urban traffic, it is a major challenge to develop prediction models that are able to generate reasonable predictions for a broad range of situations. Commonly employed data-driven approaches encounter problems related to the lack of transparency of black-box approaches and poor generalizability due to overfitting. Meanwhile, most of the publications to date have focused on the modeling part, but investigations that provide transparency into the transferability of learned patterns and the effect of different settings on generalizability are rarely addressed. This paper addresses these challenges by presenting an advanced evaluation method providing insight into the ability of models to create plausible predictions even in exceptional situations. The proposed method is applied to investigate variations in the provided input information, varying diversity in training data, and different model parameters. Among other things, our results show that providing semantic contextual information and enriching real training data with synthetic samples contributes to better generalizability. Furthermore, the evaluation revealed weaknesses of commonly used metrics, as the exclusive use of displacement errors can be misleading in terms of generalizability and plausibility of results. In summary, this contribution paves the way for reliable predictions in urban traffic by providing valuable insights and a methodology for a critical evaluation of prediction models.

Sustainability Assessment is an essential process that guides the decision-makers to the most sustainable option. In the case of the Philippine transportation sector, one of the rising transportation systems is the different ride-hailing services. The main objective of this study is to evaluate the overall sustainability of the taxi service operation in Manila and propose improvements for the system’s development. This assessment is performed using a modified Fuzzy Evaluation for Life Cycle Integrated Sustainability Assessment (FELICITA). Sustainability indicators are identified as the basis for a sustainable transport system. The data gathered is prepared as inputs to undergo the fuzzy inference system. Based on the threshold values from existing literature/standards, MC taxi, RH taxi, and traditional taxi are not sustainable. The main weakness of the MC taxi service lies in the social aspect, while the main weakness of the RH and traditional taxi is in the economic aspect.

Combustion vehicle emissions contribute to poor air quality and release greenhouse gases into the atmosphere, and vehicle pollution has been associated with numerous adverse health effects. Roadways with extensive waiting and/or passenger drop-off, such as schools and hospital drop-off zones, can result in a high incidence and density of idling vehicles. This can produce micro-climates of increased vehicle pollution. Thus, the detection of idling vehicles can be helpful in monitoring and responding to unnecessary idling and be integrated into real-time or off-line systems to address the resulting pollution. In this paper, we present a real-time, dynamic vehicle idling detection algorithm. The proposed idle detection algorithm and notification rely on an algorithm to detect these idling vehicles. The proposed method relies on a multisensor, audio-visual, machine-learning workflow to detect idling vehicles visually under three conditions: moving, static with the engine on, and static with the engine off. The visual vehicle motion detector is built in the first stage, and then a contrastive-learning-based latent space is trained for classifying static vehicle engine sound. We test our system in real-time at a hospital drop-off point in Salt Lake City. This in situ dataset was collected and annotated, and it includes vehicles of varying models and types. The experiments show that the method can detect engine switching on or off instantly and achieves 71.02 average precision (AP) for idle detection and 91.06 for engine off detection.

This paper presents an online optimization method for metro network train scheduling and passenger flow assignment based on multi-agent reinforcement learning, aiming at minimizing traction energy consumption and average passenger waiting time. The problem is modeled as a multi-agent Markov decision process using a multi-agent actor-critic framework for network train scheduling and a deep deterministic policy gradient framework for passenger flow assignment. All agents interact with the same metro simulation environment, which generates train timetables and passenger flow assignments that meet complex constraints. Results of the case study on anonymized data of Chongqing Metro show that the proposed method outperforms baseline scenarios and is able to adjust train schedules and passenger flow assignments in real-time when passenger flow distribution fluctuates, demonstrating its effectiveness and robustness.

The effective monitoring of urban traffic can be successfully achieved through the use of fixed sensors based on inductive loop detectors. These devices provide valuable information about the intensity of vehicles traversing a specific street in a particular direction. This kind of data affords a comprehensive understanding of the city’s mobility and traffic conditions. The primary aim of this study is to evaluate the spatiotemporal patterns of vehicular mobility in the city of Palma. Palma, the focus of this study, is a Mediterranean city located on the island of Mallorca (Spain). Palma has an approximate population of 400,000 residents. The city’s economy is heavily reliant on tourism, with around five million tourists visiting annually. Spatiotemporal traffic dynamics were analyzed at six monitoring stations for the period 2003-2022 located in high, medium, and low-income residential areas. The results show a significant decrease in the total number of vehicles in all neighbourhoods. Daily, weekly, monthly, and yearly mobility patterns were examined, generally showing a substantial drop in the number of vehicles. Apparent causes behind this include the development of restrictive private vehicle mobility policies, the increase in bike lanes, the reduction in the number of lanes on main roads, and the delineation of no-traffic zones. These results allow for optimism for the future of vehicle traffic in Palma, in favour of a more sustainable city. Despite the decrease experienced in vehicle counts throughout the period analyzed for the selected sensors located in residential areas, the obtained results should be interpreted very cautiously, as this situation cannot be generalized to the rest of Palma’s areas. There is evidence that traffic on certain city roads as Vía de Cintura has significantly increased.

As the rapid advancement of artificial intelligence (AI), information and communication technologies, autonomous driving system (ADS) has increased permeation into the traditional automotive industry in recent years. To reduce the Safety of the Intended Functionality (SOTIF) risk of autonomous driving system hence improving its dependability, SIL simulations are extensively exploited as virtual mileage test in compensation of the prohibitively expensive and inefficient road test. In SIL simulation, unprotect left-turn is an intricate traffic scenario to be reproduced due to the intensive interaction between vehicles at the intersection. However, most state-of-the-art commercial simulation software omit the interaction modeling. Thus, in this paper, we proposed a driver behavior modeling approach at unprotected left-turn scenarios to enhance the authenticity of SIL simulation. The left-turn scenario was modelled through three stages, including interaction selection, interaction decision and driver behavior modeling, of which a logit model and intelligent driver model (IDM) were used for the latter two stages. After model calibration, it proves this approach can generate highly authentic traffic flow with unbiased feature distribution towards the real-world, indicating its potential in SIL simulation performance improvement.

In order to identify the key influencing factors of emergency logistics system reliability, improve the reliability of emergency logistics system and ensure the smooth and efficient operation of emergency activities, Firstly, based on the connotation of the reliability of emergency logistics system, the factors impacting the reliability of emergency logistics system are organized and summarized, screened and analyzed, and the reliability assessment metrics system of emergency logistics system containing four primary indicators and seventeen secondary indicators is constructed. Secondly, the DEMATEL method was used to clarify the key influencing factors of the reliability of emergency logistics systems and the degree of interaction between each influencing factor, and use the AISM method to reveal the interaction relationships between factors. Finally, through the analysis of the model construction results, it is found that: the causal attributes of the factors determined by the DEMATEL method and the AISM method are consistent, and the hierarchy in which the factors are classified by the AISM method is significantly correlated with the causal degree of the factors obtained by the DEMATEL method. Combining the DEMATEL method and the AISM method, it was finally determined that the focus of the reliability improvement of the emergency logistics system should be placed on six key influencing factors: organizational coordination capability, rapid response capability, support of advanced technology, expert consultants, satisfiability of emergency supplies reserve, and flexibility of production system.

Taiwan is a mountainous, steep-terrain island surrounded by oceans on all sides. Even though annual precipitation totals are typically relatively high, most of that water ultimately ends up in the sea, making storing water resources difficult. The question of how to keep water through artificial constructions is crucial. The reservoir is an essential construction for storing water resources in Taiwan. Most of them obtain water supply from rivers further upstream. Whenever there is a sudden downpour, rivers carry a large amount of mud and rocks into the reservoir, which can cause sedimentation at the reservoir’s bottom and a decrease in its capacity to store water, reduced. With the Shihmen Reservoir as an example, the underwater terrain of the reservoir is depicted using the multibeam echosounder and side-scan sonar system. In addition, the underwater terrain is given coordinates by combing the global positioning system (GPS). The point cloud data of the measured results above were substituted into the geographic information system to generate a 3D digital image of the dam’s underwater structure and siltation status. Through years of data collection, the gathered information can be utilized to compare the fluctuations of silt accumulation in Shimen Reservoir over time.

This paper examines recent research in the field of public transportation, specifically focusing on the development of learning algorithms for predicting the behavior of trains and buses. However, it underscores the overlooked significance of having a clear and structured representation of data entities. To address this oversight, a relational model is proposed that captures the essential data fields specific to the subway system to enhance the learning process. The model undergoes validation through collaboration with a metro control center and domain experts. Furthermore, the study integrates this relational model into a hybrid approach that combines online and offline machine learning techniques. This approach effectively forecasts delays and passenger flow, thereby enabling informed decision-making and optimizing rail operations through a decision support system. The paper concludes by emphasizing the pivotal role of the proposed model in facilitating the selection of relevant variables for each learning problem.

We examine the problem of forecasting the spatial extent of a just-occurred traffic incident’s impact and the travel delay induced by it at certain future time points. We present and evaluate a machine learning-based solution for the above problem. The proposed solution is based on a standard classification model that takes in a variety of input features that include the incident attributes and features derived from traffic sensor data. We evaluate several versions of the solution by varying the classification model, the number of impact classes, the type of training data, and the time at which the prediction is made. This is done by conducting a series of experiments using a real-world traffic incident dataset along with the corresponding traffic sensor data. In particular, we investigate the issue of class imbalance in the incident dataset, the disparity in the class-wise prediction accuracies, the benefit of taking the incident’s early impact into account, and the relative importance of the input features. The findings of this study are potentially insightful to practitioners and researchers in the field of intelligent traffic management.

This study evaluates four methods to estimate the level of service for third-class highways through an Analytic Hierarchy Process (AHP) comparative matrix. The four methods are: i) Speed and Intensity Relation Method, ii) Highway Capacity Manual, iii) Invias Method, and iv) Percent Time-Following. A field inspection was performed to obtain highway data, such as the geometric characteristics of the road, vehicle volume, and vehicle speed. Considering this data and the corresponding methodologies of each method, the level of service of the highway is obtained. Results indicate a level of service of D in each method. The AHP comparative matrix is used to decide which method is the most adequate based on precision, time that takes to obtain the result (speed of response) and cost. Results indicate that the most appropriate method for estimating the level of service is the speed intensity relation method, mainly due to the low data required to estimate the LOS, making it a low-cost, and low-effort method to be conducted.