ABSTRACT. Despite the recent advancements of autonomous vehicle technology, performing lane changes and merging in dense traffic environments remains an open challenge. An important driving task is to find a suitable space to merge into without placing any vehicle in a collision prone situation. While humans often put themselves at risk for brief periods of time, autonomous vehicles cannot do the same by design due to obvious safety and liability issues.

In this talk, we discuss how safe is an intervehicle spacing and propose a cooperative lane change approach that relies on vehicle connectivity to achieve safe merging or lane change under different traffic conditions. The proposed approach requires that the merging vehicle negotiates the creation of a safety gap in the destination lane and, till the lane change maneuver is completed, it operates as having two possible leaders, one in its own lane and one in the destination lane. Moreover, the future following vehicle in the destination lane operates as if the merging vehicle has already changed lanes. This solution leads to a smooth creation of spacings for the vehicle to merge into while safety is guaranteed. Once the gaps are created, the vehicle performs the lane change maneuver using a proposed robust combined longitudinal and lateral controller. Simulations are used to demonstrate the safe lane change maneuvers under different traffic conditions.

ABSTRACT. This paper presents a literature review on the application of traffic simulation for the evaluation of traffic safety levels. The main aim is to identify, through the implementation of a multi-step methodology current research-trends, main gaps in the literature and possible future challenges. First, a bibliometric analysis is carried out to obtain a broad overview of the topic of interest. Subsequently, the most influential contributions are analysed in-depth, with specific attention to specific issues.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Navigation safety is a priority both at European and global level. Despite the important progress made over the years, sea accidents remain a major concern and much work is still needed to enhance maritime safety. Knowing the causes and precursors of past accidents is essential to identify the elements on which to intervene to improve safety and reduce the possibility of an accident to occur again. In this study, 1.079 sea accidents from the IMO database are analyzed using Semi-supervised Recursively Partitioned Mixture Models in an attempt to identify and categorize causal themes from accident data. Special attention is devoted to the human element, which is widely recognized as a primary or precursory cause in most accidents.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Road crashes have historically plagued the Gulf Cooperation Council (GCC) countries. Unfortunately, vehicle crash studies conducted in the GCC region have been scarce, making it difficult for decision-makers and researchers to assess the magnitude of the road safety problem regionally and to tackle it effectively. In the present study, the authors use a multivariate logistic regression model to investigate the contributing factors to fatal road crash severity in the Emirate of Abu Dhabi, part of the United Arab Emirates (UAE). Data was collected from all reported crashes occurring between 2012 and 2017. This study may be relevant not only to the UAE, but also to other GCC countries (i.e., Saudi Arabia, Qatar, Kuwait, Oman, and Bahrain) due to their similarities in road design, vehicle fleet, and driving culture. Based on study findings, the authors recommend Abu Dhabi to focus on: increased law enforcement during late hours, as well as in areas where and during times when impaired drivers may be more likely to be caught; increased use of technology capable of mitigating the negative effects associated with dusty and foggy weather conditions; safe pedestrian-oriented road design and transport policies; improved safety and design standards on higher-speed-limit roads; implementation of safety measures capable of preventing running-red-light events; effective educational campaigns and training programs in changing driving culture, especially among the Emirati residents. The authors also point out the need for change in traffic legislation in order to more effectively address impaired driving/reckless behaviour (e.g., tailgating), so that violators are punished more proportionally to the damage they may actually cause to society. Punishment measures could entail jail time and/or license cancelation.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Intersection crashes have accounted for a large portion of all fatal road crashes worldwide. Thus, much research has been devoted to comparing safety levels among different intersection types, such as signalized intersections and roundabouts. While previous research efforts have been insightful, none have focused on crashes occurring in the United Arab Emirates (UAE). This is worth noting because while several states in the United States (US), as well as countries in Europe, have replaced many of their signalized intersections with roundabouts as an attempt to curb severe injuries and improve safety, the UAE appears to be following the opposite trend (e.g., replacing many of its long-standing roundabouts with signalized intersections). Hence, the objective of this research is to compare the safety performance of signalized intersections and roundabouts in the Emirate of Abu Dhabi, the largest Emirate in the UAE. The study collected 6 years’ worth of road crash data. Crash severity levels between the two intersection types were compared, while controlling for factors such as posted speed limit, number of vehicles involved in the crash, and vehicle type. There was no significant crash-severity-level difference between the two types of intersection designs studied, suggesting that their safety performances are comparable. The authors discuss how future research could focus on identifying the factors that might have contributed to this finding differ from studies conducted in the past.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. This study analyzes and compares reaction times of motorists at the onset of a yellow traffic light under manual and automated driving, based on experiments performed using a driving simulator. Results show that reaction times of subjects driving an automated vehicle that experiences a failure when approaching a signalized intersection are higher than those of subjects driving manually. When the analysis is restricted to automated driving, results indicate significant differences between reaction times at the first system failure and those at subsequent ones.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The extraction of travel time data from Bluetooth is nowadays widely used in various countries. However, data from several studies shows a large variation in the equipment rate with Bluetooth devices in different areas of the world. In particular, most research points to significantly higher rates in European countries in comparison to the USA. Based on the evaluation of a dataset of almost two billion single Bluetooth detections, this work suggests one main reason for this difference being the significantly higher Bluetooth equipment rate amongst trucks on German freeways in relation to passenger cars. The findings show that trucks have an approximately five times higher chance to be equipped with a detectable Bluetooth device.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The purpose of the paper is to determine and examine, to what extent blockchain scenarios for shipping industry have practical explication from maritime ports’ perspective and how these are sync with ports’ long-term development strategies, particularly in Denmark. The present study involved qualitative interviews with representatives of the biggest maritime ports of Denmark, varied by location, volumes, operations and cargo type. Data saturation is achieved through several round of in-depth semi-structures interviews. Results showed uncertainties in long-term investment strategy of the considered ports. While focused on land expansion and operation development, the port authorities lack inner-port coordination with related enterprises, which consequently affects overall efficiency. While the development strategy appears to be identical among the port authorities, it varies significantly within specific blockchain scenarios and port’s strategy regarding short-term port optimization. Besides, the role of port authority was debated. Authorities are willing to be more involved in supply chain operations as consultancy rather than just a controlling party, yet are burden by the state restrictions. Unlike generally-discussed blockchain compatibility studies, the current research contributes by revealing core business uncertainties within port area development and communication. Moreover, the case could serve as a representation of small- to middle-size ports in the EU.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. A promising approach to avoid congestion is to distribute traffic in the street network in a system-optimal manner. This could be calculated by a centralized traffic management system to be communicated as route advice. However, human drivers can independently decide whether to follow the route advice or to dissent. When trying to influence route choice, drivers’ travel experiences may influence their reaction to route advice. When recalling past events, people tend to remember the extremes (best and worst of times) more easily than the most typical of time (called impact bias) and if unaware of the atypicality of an event, drivers may unconsciously rely on it when making an affective forecast of a similar future event and act accordingly (affective forecasting). To investigate if this behavior also occurs in route choice decisions, a conceptual replication of the study of Morewedge, Gilbert, and Wilson (2005) was made. Using an online-survey, drivers (n = 155) were asked to recall and rate their affective memories of past congestion events (four recall conditions, between-subject design) and to forecast their affective reaction to an upcoming congestion on their next trip. Further, they had to make route choice decisions, given route advice with varying congestion share on the route alternatives. Results show that drivers indeed tend to recall extreme experiences. However, the affective quality of the event recalled could neither explain the affective forecast of upcoming events nor the driver’s route choice decision. Exploratively, it was found that drivers were generally more willing to face congestion than to consider bypassing it by a detour.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Abstract Inter-urban roads in Indonesia accommodate heterogeneous traffic with mixed types of vehicles sharing the same lanes. The heterogeneity results in varied average speeds and speed deviations of different vehicle categories. Lack of posted speed limits, misplaced speed limit posts, and the vehicle speed and acceleration performances are among the possible factors leading to a wider spectrum of speed variations. Understanding the speed characteristics of heterogeneous traffic is essential in traffic safety improvement. Previous studies have been conducted to investigate speed characteristics but few studies concentrate on the effects of heterogeneity. This study analyzes the speed characteristics of traffic with 6 vehicle categories along inter-urban roads of 19 road sections in 8 provinces in Indonesia. Within-group and between-group analyses were conducted on speeds and speed variations, and geographical comparisons by vehicle categories were made to a reference province. As earlier studies indicated that speed and speed variation contributes to accident occurrences and severity, the results of this study can contribute to traffic policing, setting up the speed limits and traffic safety improvement programs.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Several new concepts and innovative technologies emerged in the last decade to overcome the problem of urbanization, which is hardly satisfiable with private vehicles or conventional public transport services alone. One of these is the Mobility-as-a-Service (MaaS) concept. MaaS is a user-centric, intelligent mobility distribution model, in which user needs are met via a single platform and transport options are offered by a service provider (MaaS operator). The framework was realized in the MaaS4EU project, which main goal is to provide quantifiable evidence, frameworks, and tools, to remove the barriers with the help of the MaaS concept. The project provides insights about MaaS costs and benefits in three complementary pilot cases. One of these pilots started in Budapest in late 2019 involving several mobility service providers, such as public transport, bike-sharing, car-sharing and taxi. The pilot demonstrates the business case when a private service provider is the MaaS operator. This article provides information about the living lab of this pilot. It presents the used MaaS approach, the involved stakeholders, the level of integration, and the proposed solutions, but most importantly it shares the lessons learnt from the pilot and delivers results of a real-life demonstration.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. A reinforcement learning-based dynamic congestion pricing method for morning commute problems is proposed. In this method, tolls are iteratively updated day by day based on observable information such as traffic volume. The advantage of the method is twofold. First, the method does not require traveler's personal preferences such as value of time. Second, the method does not require manually designed toll update scheme; instead, the method automatically find the most efficient toll by a data-driven manner. Results of numerical experiments show that the proposed method properly decreased traffic congestion in various types of networks.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Within the following decade, the wide deployment of Vehicle Automation and Communication Systems (VACS) is expected to influence traffic performance on motorways (Markantonakis et al., 2019). Apart from safety and comfort, if exploited appropriately, the emerging VACS may enable sensible novel traffic management actions aiming at mitigating traffic congestion and its detrimental implications. This study investigates the use of some proper traffic control strategies for mixed vehicle traffic. These strategies include time-gap adaptation, variable speed limits, lane changing control actions, utilizing VACS in different penetration rates, and are tested within a microscopic simulation environment using realistic demand profiles and appropriate model parameters for connected automated vehicles. Preliminary results demonstrate significant improvements even for low penetration rates.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Traffic congestion in big cities has been proven to be a difficult problem with suboptimal effects in terms of driver delay and frustration, cost and impact on the environment.~In principle, many transportation networks lack a unified framework, which will coordinate the traffic in such a manner, in order to suppress congestion and at the same time improve the travel time of the users situated in it.~The rapid advancements in information, communication, and computation technologies have given rise to more elaborate modeling frameworks, aiming to act as the coordination unit necessary to counter the issue of congestion in real-time conditions.~Nevertheless, most drivers situated in a transportation network prefer to improve their own travel time~(user optimum) and tend to disregard the overall performance (collective optimum) of the network~\cite{Colak2016}.~Such actions might have an adverse effect on the efficiency of the network, prospectively leading to greater waiting time intervals for each individual driver.~We propose a macroscopic model equipped with an underlying reservation feature, known as Route Reservation Architecture~(RRA).~Vehicles enter the network~(mainstream-wise or from the on-ramps) as long as this inflow does not incur a density that exceeds the network's critical density.~Those vehicles prospectively exceeding the critical density are stored as queues at the origin of the motorway stretch and within the on-ramps~\cite{Kotsialos2004}.~Once there is sufficient space, for those vehicles to be accommodated by the respective cell of the network, they are discharged from their queueing instance at their respective origin, moving towards their assigned path freely.~In previous works, this architecture has been only applied in microscopic simulations in the context of urban networks without the influence of source terms~(on-ramps)~\cite{Menelaou2017}.~When the critical density of the stretch is reached, the reservations are activated, instigating a waiting interval to the vehicles stored at queues, allowing vehicles only to enter the network at a later time instant, such that the critical density is not crossed.~In this vein, we avoid the congested region and operate only at free-flow conditions.~Our target is to investigate the effectiveness of this architecture, i.e. to minimize the Total Travel Time (TTT) of the vehicles present in the network, along with the vehicles situated in the queues, through a macroscopic simulation framework.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. This work aims to examine the opportunities of a pricing mechanism applied along with the route-reservation architecture. Route-reservations have shown to be an efficient traffic-congestion alleviation mechanism that navigates drivers to their destination at the earliest possible time. This is achieved through the use of a novel route-reservation mechanism responsible for regulating the routes and the departure times of drivers in an effort to sustain the density of vehicles below the network's critical value. In doing so, the overcrowded conditions will be avoided with the increasing demand shifted in both space and time dimensions. Nevertheless, the improvements of route-reservations are justified, only assuming that the majority of drivers adhere to reservation schedules. In this work, we remove this assumption by enabling drivers to depart at the time that their utility is maximized.

The main component of the architecture is the central scheduler that is responsible for providing all navigation instructions to drivers, monitoring the utilization of each road segment, and making the appropriate route-reservations. In this setting, when a driver is ready to start his/her journey communicates to the central scheduler, his/her O-D pair and the time that desires to depart. In turn, the scheduler responds with the path that the driver should follow and also suggests the appropriate departure time. The provided path and departure time ensure that a driver will not traverse through the congested parts of the network so as to arrive at the destination at the earliest possible time. In turn, the driver is responsible for taking the provided path while also informing the scheduler about his/her departure time choice. Finally, based on the driver's departure time choice, the scheduler makes the appropriate reservations at the time-slots that a driver is expected to traverse each road-segment of the assigned path.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Environmental pressure and technological development is finally bringing Battery operated Electric Vehicles (BEVs) to the roads. Continuous efforts are made to mitigate vehicular emissions by introducing BEVs as an alternative to the Internal Combustion Engine Vehicles (ICEVs), due to zero exhaust emissions and higher efficiency of electric motor. Numerous studies have shown the energy-efficiency benefits of BEVs with respect to ICEVs, but not when BEVs and ICEVs co-exist. In this paper, we evaluate the overall energy-efficiency of mixed BEVs-ICEVs scenarios in-terms of total gasoline and electricity consumption for various penetration rates of BEVs at an isolated intersection. We employed our intelligent intersection management architecture (IIMA) equipped with the synchronous intersection management protocol (SIMP) against the conventional Round-Robin (RR) intersection management. Simulation results using the SUMO simulation framework show that SIMP outperforms RR for different penetration rates of BEVs, achieving up to ~200% upturn under particular traffic conditions.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Referring to the annual accident report (2018) of the European Commission it is clear that during the decade 2007-2016 fatalities by mode of transport have dropped significantly inside the European Union (EU). However, there’s always a place for further improvements. It is also clear that the changes in the socio-economic and demographic context of the EU countries have reported impacts on the fatalities by different modes of transport and overall to their road safety levels. Therefore, in order to support the road safety strategies of the EU countries that have an under-performing system (in terms of road safety levels), it is essential to incorporate these socio-economic and demographic factors and provide a comparative analysis of the performance of each EU country. This paper aims at investigating the road safety performance of 30 EU countries over the time period 2007-2017 by incorporating their socio-economic and demographic context. Additionally, this paper will provide interesting findings concerning the recorded fatalities for the different modes of transport. Furthermore, the comparison between the EU countries’ performance will be provided from the development of a Benchmarking Analysis (extended form of Data Envelopment Analysis). The resulted technical efficiency scores of the countries with their socio-economic and demographic context will expressed using censored regression, i.e., Tobit regression analysis. By this we will identify the components that effect the most of the countries’ road safety performance. In detail, this paper will provide the overall ‘picture’ of the road safety level inside the EU region.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Presented herein is a model for the optimization of dynamic maintenance scheduling in bridge networks, considering the condition of bridge elements and the vulnerability of bridges. Thus, the proposed model focuses on multi-system multi-component networks since a bridge is considered as a system of components.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Real-time conflict prediction models (RTConfPM) are an innovative approach to deal with real-time road safety analysis, even in absence of reliable crash data. This paper presents a RTConfPM to predict rear-end crashes, using Time-To-Collision (TTC) values recorded with radar sensors on multiple motorway cross-sections to define unsafe situations, and traffic conditions recorded on the same sections as inputs to the model. Several classifiers were trained and compared: K-Nearest Neighbors (KNN), Naïve Bayes (NB), Discriminant Analysis (DA), Decision Trees (DT) and Support Vector Machine (SVM). All the proposed models provide better performance than most of crash-based real-time models found in the literature, with KNN and SVM significantly better than the others when considering Recall indicator.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. One of the main systems for ground vehicles is the Electric Power Assisted Steering (EPAS). Due to its importance, a redundant system is usually implemented. This solution remain however expensive, especially for small urban cars. This paper proposes an alternative fail-safe control algorithm using this time differential braking. Controlling each brake apart generates a yaw moment enabling turning the vehicle. Optimal coordination is ensured due to the importance of this maneuver. Results show that differential braking, that already exist in most of passenger cars, is able to generate a sufficient additional yaw moment to help the driver steer his/her vehicle. This can reduce the space and the cost of the vehicle.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. On specific congested sites of cities, drivers, attempting to minimize their own travel time, use secondary roads instead of the main roads. This rat-running behaviour may affect network performance, but little research has been conducted in this topic. The present study explores rat-running network impacts and suggests evaluation of network performance in a global perspective with respect to travel time, pollutant and noise emissions, and also road safety. For this purpose, vehicle dynamics data need to be collected, which is relatively easy to obtain through GPS-based traffic data. A case study involving a secondary residential road with intersections, high slope and narrow sections was selected for testing the proposed approach. Different scenarios of mitigating measures that could be implemented within the study area to reduce the number of rat-running drivers (rat-runners) are explored using the traffic simulation modelling software VISSIM. Results show implementing a single network intervention/redesign measure at local level may minimize externalities, but may not be sufficient. This study will allow traffic planners and designers to rethink certain measures to reduce the number of rat-runners moving through specific areas.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Despite the worldwide growth of low cost carriers focusing on non-stop, point-to-point routes, many air travelers still change planes, especially on long-hauls. The volumes and shares of transfer passengers at different geographic levels can make useful air transport indicators for a number of industry, policy and research questions. For example, actual and future transfer passenger numbers are likely to be a major driver of airside non-aeronautical revenues and, hence, may have to be taken into account when designing new or converting existing terminals. From a competition policy perspective, e.g. in the context of airport market power assessments, transfer passengers might have to be treated differently than local ones, as their relevant markets differ. And noise-affected residents in the vicinity of airports may want to point at (high) transfer passenger shares when questioning airport expansion plans, arguing against additional “import of noise”.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. In this paper, the authors investigate the market potential for long-haul narrowbody aircraft using Sabre’s Market Intelligence origin-destination demand and Innovata’s flight schedules data. The origin-destination demand data includes information on the real itinerary of passengers, starting with the first airport of embarkation and ending with the final destination of the trip. Moreover, information on the routing and airlines is provided. Hence, this data typically provides more information than official statistics, which typically report only on the number of passengers on board a flight between two airports, without indicating the true origins and destinations. Hence, the Sabre MI data allows estimating the real demand potential on a particular city pair. This data is matched with flight schedules in order to identify whether non-stop flight connections exist. With this approach, airport pairs without non-stop connections, but with a potentially sufficient demand level, can be identified by a gravity model-type approach and the probability of airlines offering new routes can be estimated with a logistic regression.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. In autumn 2017, Air Berlin, the second biggest player in the German airline market, discontinued its operations due tobankruptcy. Prior to this, the struggling carrier had already been reducing its domestic capacities and adjusting its servicepatterns, which had impacts on the competitive situation for the remaining carriers on various domestic origin-destination (O&D)markets. This paper aims at identifying resulting changes of, e.g., the yields generated in intra-German air transport. In theGerman domestic market, the relevant competition per route can be narrowed down to a few players as opposed to continental orintercontinental flights. We found evidence that on 90% of the TOP20 O&Ds, representing 96% of the total passenger volume,revenue per pax (as an indicator for yield) increased. We found FRA-BER and MUC-BER – both O&Ds with intense Easyjetcompetition to be the only O&Ds that did not show an increase in price/yield after the market exit of Air Berlin.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. In the last years, scientists as well as political activists proposed the introduction of ambitious prices for all CO2 emitting sectors, including aviation. Suggested prices range from about 45 - 350 €/ton of CO2. Such prices are considered an indispensable element of a strategy aiming at stabilizing the global temperature increase well under 2.0 degrees Celsius (Paris objective). For comparison: As of January 2020, the price for European Emission Allowances in the European Emission Trading Scheme was about 25 € per ton of CO2. European Air Transport has been participating in this scheme on a mandatory basis since 2012.

How would such ambitious CO2 prices affect air transport? How likely are a significant increase in airfares and a corresponding decrease in demand? This paper investigates the potential impacts of high CO2 prices on airfares and growth in aviation. For this, we analyze the relevant literature and conduct some model-based estimations. In addition, we provide a rough estimate of the economic impacts in case not only CO2 but all climate relevant species from aviation (NOx, SOx, H2O, aerosols, contrails and contrail cirrus) were subject to emission pricing.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The enormous economic impact of air transportation network on local, national and international level has created the interest for further investments and thus creating more complex air transport networks. The increased number of worldwide airport passengers and of the aircraft movements and also of the international airport cargo shipments is evident. Therefore, the functionality of this complex air transport network is very important and requires to be investigated and evaluated for identifying the airports that appear to be critical to this network. However, besides the economic benefits that air transport network are offering in local, national and international level through their services (shipments of goods, transport of passengers) they play an ever-increasing role in safeguarding global health security. This paper investigated the global air transport network for identifying the airports that may constitute a public health event of international concern from infectious diseases to those related to food safety etc. In detail, critical airports, in terms of Betweenness centrality, Degree centrality and Closeness centrality, were identified and addressed by pointing them to global authorities for suggesting and implementing routine prevention and control measures for a possible future disease outbreak.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Conventional intersection management that allows multiple vehicles from one road at a time, e.g., Round-Robin (RR), may constitute bottlenecks in urban traffic management. Consequently, new intelligent intersection management (IIM) approaches were proposed to reduce time loss, fuel wastage, and ecological damage. IIM is also suited to take advantage of the new communication capabilities of autonomous vehicles that are emerging, though still co-existing with human-driven vehicles. This paper extends the analysis of a recently proposed synchronous IIM system, the Synchronous Intersection Management Protocol (SIMP), that is compared with the RR scheme in a four-way single-lane intersection as those found in urban residential areas, under maximum vehicle speeds of 30km/h and 50km/h and various traffic arrival rates. We characterize performance by measuring time loss, i.e., the additional trip delay due to forced slowdown, supported with an analysis of fuel efficiency. The experimental results obtained with the SUMO simulation framework indicate an advantage for SIMP in both metrics, which becomes more significant under low traffic intensity and low maximum speed values.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The following paper analyzes the operation of autonomous vehicles (AVs) in a private and shared scenario as well the influence of adding a toll in the main bridge that connects a residential area to the city center, representing a case study of the city of Florianópolis in Brazil. We develop a mathematical model whereby the aim is to minimize the overall generalized costs and evaluate its effect on the congestion problem. The main aim of this paper is to estimate some of the impacts that can come up from the deployment of AVs in this urban area where costs of operation, parking and toll are considered and minimized. Several penetration rates of AVs with different shares of private-owned and shared vehicles are analyzed together with the influence of a toll that affects the trips distribution and congestion depicted by the level of service. The main conclusion is that effects on congestion, that can be depicted through the level of service, will only be noticeable when a toll is added and AVs represent a vast majority of the vehicle fleet and some of them are not-private owned, i.e., are shared vehicles. A level of service A will likely be obtained once there is a penetration rate of 50% and 80% of these AVs are shared. When AVs are 80% of the vehicle fleet, 50% of these shall be used for shared purposes. These results can only be obtained if a toll is added to the bridge, that will increase the travel costs and likely force passengers to change their travel behavior.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. A cooperative framework based on an MPC-based path-planning algorithm for automated vehicles on multi-lane motorways has been designed. A number of connected vehicles, which are able to communicate with each other (vehicle-to-vehicle (V2V)), or the infrastructure (vehicle-to-infrastructure (V2I)), in order to send or receive real time information about their path (position, lane, speed, acceleration) or the paths of the other vehicles, are guided by the path-planning algorithm. To account for various uncertainties, the approach is cast within a model predictive control (MPC) framework, which is real-time feasible thanks to low computation times. The proposed approach is embedded within the Aimsun micro-simulation platform allowing the investigation of the impact of the suggested approach on the traffic flow as a whole for increased penetration rates of connected-automated vehicles. Preliminary results indicate that the proposed framework enhances not only the efficiency of the optimally controlled connected vehicles, but also the performance of the whole section, for higher penetration rates of connected vehicles.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Efficient utilisation of urban road networks has been in the epicentre of researchers’ attention for many decades already. Nowadays, many cities rely on traffic-responsive strategies for the effective management of traffic flows. However, state of the art methodologies, based on Artificial Intelligence (AI), have started challenging the currently implemented solutions. Reinforcement Learning (RL) stands as one of the most promising AI-based methodologies aiming at the optimization of traffic signal controlling. Since the first appearance of relevant RL methodologies, numerous implementation strategies have been suggested. Nonetheless, these approaches are usually evaluated under ideal or simplistic conditions (e.g. toy networks, unrealistic demand patterns, etc.) thus lack the ability to assess the effectiveness of RL-based signalling optimisation under realistic conditions. The currently presented study bridges this gap by evaluating a plethora of RL-based implementations under fully realistic urban demand conditions as manifested in the large-scale road network of Nicosia, Cyprus.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Automated vehicle technology can be beneficial for many aspects of transport, especially, improving traffic flow stability and efficiency. However, the influence of different automated driving styles on traffic efficiency is still not fully understood. Transport systems are very complex and non-linear, i.e. many participants with different characteristics interact with each other and the aggregated result of their interactions could cause a remarkable change in the entire network. Considering that automated vehicles with different driving styles interact with the environment in different ways. In this study, we try to understand the influence of different automated driving styles (e.g., cautious, normal, aggressive) on the important variables in traffic flow theory (e.g., speed) to reveal their impact on network efficiency. Characteristics of these driving styles are extracted by clustering the highD dataset and then, translated into different car-following models for simulation in the SUMO traffic simulator environment. Multiple scenarios of mixed traffic conditions (i.e. ranging different ratios of driving styles) are simulated on the network of Munich inner city.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The proliferation of next generation mobility, promotes the use of autonomous cars, connected vehicles and electromobility. It creates novel attack surfaces for high impact cyberattacks affecting the society. Addressing the cybersecurity challenges introduced by modern vehicles requires a proactive and multi-faceted approach combining techniques originating from various domains of ICT. Emerging technologies such as 5G, LiDAR, novel in-vehicle and roadside sensors and smart charging, used in modern cars, introduce new challenges and potential security gaps in the next generation mobility ecosystem. Thus, it is critical that the domain’s cybersecurity must be approached in a structured manner from a multi-domain and multi-technology perspective. The CARAMEL H2020 project aims to address the cybersecurity challenges on the pillars upon which the next generation mobility is constructed (i.e., autonomous mobility, connected mobility, electromobility). To achieve that, advanced Artificial Intelligence (AI) and Machine Learning (ML) techniques will be utilized for the identification of anomalies and the classification of incoming signals indicating a cyber-attack or a cybersecurity risk. Apart from risk detection, methods for the mitigation of the identified risks will also be continuously incorporated to the CARAMEL solution. The final goal of CARAMEL is to create an anti-hacking platform for the European automotive cybersecurity and to demonstrate its value through extensive attack and penetration scenarios. In this paper we will expand on the unique cybersecurity-relevant characteristics of the pillars upon which the CARAMEL solution is built. Next, a number of use cases emerging from such analysis will be extracted in order to form the basis upon which the CARAMEL platform will be evaluated. Finally, we will conclude with an overview of the platform’s architectural composition.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Intelligent Transport Systems (ITS) have emerged as an integral part of smart cities, providing increased ease of mobility as well as efficiency and safety in vehicular traffic. Given its wide array of applications, ITS has also become a multidisciplinary field of work where vehicular communications, traffic control, ADAS (Advance Driver Assistance System) sensors, and vehicle dynamics have all to be accounted for. The study of such diverse aspects makes the evaluation of new ITS approaches, algorithms, and protocols not a small feat. For this reason, the availability of an effective, scalable, and comprehensive tool for the investigation and virtual validation of new ITS solutions is paramount. In this work, we present a simulation framework, called CoMoVe (Communication, Mobility, Vehicle dynamics), that effectively addresses the above need, as it enables the virtual validation of innovative solutions for vehicles that are both connected and equipped with ADAS sensors. Our framework encapsulates the important attributes of vehicle communication, road traffic, and dynamics into a single environment, by combining the strengths of different simulators. CoMoVe finds its use to evaluate the impact of vehicle connectivity, while imposing causality on vehicle dynamics and mobility. Such an assessment can greatly facilitate the development of control systems, algorithms, and protocols for real-world ITS.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. In this paper, a novel framework for a benchmark system for autonomous vehicles focusing on their security and reliability is proposed. Computer vision and networking technologies are improving offering solutions towards automation in connected autonomous vehicles. These systems are using sensor technologies, including vision and communication, providing information and measurements for the environment and other connected vehicles. As a result, unlike conventional vehicles, autonomous vehicles have to communicate with other vehicles as well as other external network infrastructure. However, such requirements make autonomous vulnerable to the attack. This may also motivate various types of cyber threats and attacks like traffic signs modification, GPS spoofing, and Vehicular Adhoc network distributed denial of service. Hence, this paper explores various aspects of security issues, vulnerabilities, exploitation methods and the adverse effect of them on connected autonomous vehicles and proposes a novel benchmark framework focusing on physical and communication-based attack to evaluate and assets the state- of-the-art technologies that are currently used during cyber-attack.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Electric vehicles (EVs) gain great attention nowadays since the electrification of private and public transport has a great potential to reduce greenhouse gas emissions and mitigate oil dependency. However, the influx of a large number of electrical loads without any coordination could have adverse aects to the electrical grid. More importantly, the complexity in the coordination of a large number of EVs, pose critical challenges in ensuring overall system integrity. A typical attack found in the controllers of connected EVs is false data injection (FDI), which can be utilized to distort real energy demand and supply figures. Energy distribution requests may therefore be erroneous, which results in additional costs or more devastating hazards. The lack of a proper coordination scheme, robust to such cyber attacks could cause voltage magnitude drops and unacceptable load peaks. In this work, we study the impact of FDI attacks, on various decentralised charging protocol with reduced computational requirements. The proposed decentralised EV charging algorithms only require from each EV to solve a local problem, hence the proposed implementation require low computational resources. An extensive evaluation study highlights the strengths and weaknesses of the presented solutions which are based on iterative convex optimization solvers.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Vehicles nowadays are equipped with a vast amount of sensors that collect data for the vehicle and its environment. This, combined with the acceleration of the automotive industry towards interconnected and autonomous cars, suggests that security and specifically the ability to detect compromised nodes, collect and preserve evidence of an attack or malicious activities emerge as a priority in successfully deploying the Internet of Vehicle ecosystem. Until today Digital Forensics attempts are concerned with in vehicle forensics. In this paper we present the challenges of integrating digital forensics in an IoV ecosystem and we introduce the Attack Attribution and Forensics Readiness Tool of the nIoVe system, an integrated holistic cybersecurity solution for IoV.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. We propose an Integer Linear-Programming model based on sequencing that captures all operational constraints and maintenance requisites while operational costs are minimized and schedule changes with respect to original plans are minimized. Its mathematical formulation is based on a structure where we consider tasks to be assigned as nodes and the connections between them arcs (edges). Mathematically, we consider one type of node (we treat all tasks in the same manner, despite their nature) and one type of arcs.Due to combinatorial complexity of the problem (there are millions of binary variables in a typical real instance), the solving process becomes very hard. Exact methods and/or commercial software fail to solve real case studies in reasonable time. In order to efficiently solve the problem with the formulation we propose, we employ a rolling horizon approach to obtain an initial feasible solution to feed a Relaxation Induced Neighborhood Search, which explores a neighborhood of the current incumbent solution to try to find a new, improved incumbent.We evaluate and validate the presented approach with realistic case studies featuring more than 2000 flights and 120 aircraft. We demonstrate empirically that optimal solutions are obtained in computational times which are low enough to be compatible with near real-time operations. Results and comparisons with real scenarios show the goodness of the approach.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The paper deals with the application of two meta-heuristics for fine tuning of fuzzy logic model. Two meta-heuristics are used to determine the final set of fuzzy rules and the domain of membership functions for airline market share model. The model aims at determining the airline market share on long-haul routes originating from London airport system. For this purpose, the real data was used encompassing several long-haul routes that are characterized by competition between full-service carriers (FSCs) and at least, one low-cost carrier (LCC). The results show that the application of these two meta-heuristics can significantly contribute to improvement in the overall model performance.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. As traffic volume has been increasing in Germany due to passenger and freight transport growth, time critical delivery services are facing new challenges. A possible solution arises from the technical progress in air transport research within the last years enabling the use of automated, light aircraft, which can operate in a more flexible manner and do not produce air pollutants. In order to establish an innovative airfreight chain, airfields have to meet the applying infrastructural requirements. The aim of this work is therefore to identify an according legal framework. We examined existing, progressing and missing regulations in order to assess the technical feasibility and point out further demand for legal assurance. The investigation shows that most applying regulations are technically feasible for the proposed airfreight transport, but a regulatory framework is not yet clarified for drone aerodrome design as well as unmanned aircraft systems operations at aerodromes.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Although redundancy is often associated with lower efficiency, redundant networks can provide integrators with the flexibility and capacity to respond to issues in the delivery network. This study analyzes the air transportation network strategies and robustness of the four largest European integrators, estimating the loss in connectivity when a node becomes unavailable. Accordingly, we develop a robustness index that accounts for the importance of airports in the network to each of the four main integrators, in terms of connections and freight capacity. The outcomes reveal that integrators operating within a hub-and-spoke type network are less resistant to network disruptions, while robustness is higher in less concentrated networks. The results highlight that integrators develop different network strategies to manage the trade-off between network efficiency and robustness, which may vary depending on the standardized or customized services expected by clients.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. With the advance of intelligent transportation systems (ITSs) and data acquisition systems (DAS), it is possible to explore the determinants of urban spatial-temporal mobility patterns using multi-source heterogeneous data. This study aims to use the points-of-interests (POIs) data, house-price data, and floating car data to identify the factors influencing urban mobility in Shanghai City. Within a scale of 0.5 km grid, trip production and attraction were stratified according to the traveling intensity, and the critical information related to economy, intermodal connection, land use, and time were also obtained through the multi-source data. The experiment results from an ordinal logistic regression (OLR) analysis show that average house price has a dominating and positive effect on the traveling intensity for both trip production and attraction, followed by land-use factors including finance, healthcare, residence, living service, hotel, cultural and educational service, government, corporation, catering, as well as leisure and sports service. However, the effect of scenic spots is found significant only on trip attraction. In addition, shopping is found to insignificantly affect the traveling intensity for both trip production and attraction. Unexpectedly, time factors also have diverse impacts. These findings are expected to help better understand the relationship between urban mobility and built environment factors, providing passengers with better services, and offering useful insights into urban and transportation planning.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The quality of the information provided by such systems are validated by the comparison of observed on-site data against the prediction model estimates. Although many forms of ground truth (GT) strategies exist, there is still no scalable method that can significantly cut down data collection costs. The hypothesis being tested in this study is that parking behaviour can give us a better insight for on-site collection strategies by capturing the parking dynamics within a city. This paper examines parking behaviour dynamics within Munich by inferring from parked-in and parked-out events data from vehicles. The parking behaviour was analysed using different clustering techniques to identify similar neighbourhoods within the study area. Mainly temporal features from the parking events dataset was used together with points of interest (POI) data from OpenStreetMap. Clustering was the chosen method for identifying groups in an unsupervised learning approach to capture patterns that are difficult to identify with inherent noise in the POI data.This study is part of a research project that aims to find a scalable method to efficiently measure OSPI quality on a large scale and test the potential of the transferability of the method to different cities.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. The evaluation of service staff performance is an important factor for the optimization of service level at facilities like toll plazas. Performance of staff is as multi-attribute decision making (MADM) problem, as his/her performance gets influenced by various factors including operational, ergonomics etc. Hence, in the present study the Analytical Hierarchy Process (AHP), a MADM method is used for prioritizing the criteria for increasing the efficiency of the tollbooth operator. Based on the available literature, various factors affecting performance of tollbooth operator were found out including service time, his/her capability and safety. Tollbooth operators from different toll plaza were enquired in structured questionnaire. The weights are obtained from AHP relative importance matrix and finally these weights are used for setting the priorities. The output thus will be used for concessionaire to meet up with the requirements of tollbooth operators for enhancing their performance, so that service level of toll plaza will also improve.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Social networks are strongly present in the daily life of modern society. Most people use these social networks to share information about their lives, their opinions, places they visit and their state of mind. Generally, these posts are composed of various information, being the location of the users location part of the data. The purpose of this work is to obtain the location of the posts and observe the users mobility pattern in the city of Porto, Portugal. This paper discusses the technologies available for obtaining the data, the social networks currently worth studying and their respective restrictions. It also explores new approaches to collect the data from the desired social networks, respecting all restrictions currently applied. The different software solutions developed for the social networks interactions are explored and described in depth. Subsequently, the necessary software for social networks is reviewed, the possible algorithms for data mining are discussed and its implementation is presented. Finally, the results obtained are interpreted and studied according to the characteristics of the city, tourism promotions and transport routes. Keywords: social networks, data mining, mobility patterns, clustering, urban planning.

Download: (a) Extended Abstract (b) Presentation video

ABSTRACT. Malta is a small island state in the Mediterranean, with one of the highest population densities and per capita car ownership rates in the European Union. University campuses located in urban areas have a substantial role to play in transport planning and policies since they are amongst the largest generators and attractors of commuters. However, at the University of Malta, mobility to and from the work and study place is still characterised by high levels of private car use, resulting in traffic congestion, parking constraint and land use uptake. The current transport system is discriminatory towards those outside the car, who cannot or do not want to drive. These people find themselves struggling for space to walk and cycle, whilst bus users have to withstand long journey times due to car traffic and limited priority. A participatory Active Travel workshop was organised at the university in order to better understand the determinants of pedestrian and cyclist mobility to and from the university campus. Participants included people from the university community as well as other interested stakeholders. Four different guided walks or cycles led participants along predefined paths in the surroundings of the university, to assess the environment from the perspective of a pedestrian or cyclist. Information on the active travelers’ experiences allows the assessment and spatial distribution of perceived walkability and bikeability around campus. These outputs are then presented in a participatory web map that has evident potential to assist in the design and improvement of public space for promoting active travel. Contributions from the workshop are used to identify barriers and opportunities for walking and cycling and to construct tangible policy recommendations for promoting active travel in the vicinity of the university and the wider urban environment in Malta.

Download: (a) Extended Abstract (b) Presentation video