ABSTRACT. Encouraging active and sustainable transport modes in order to limit the excessive use of cars, reduce pollutant emissions and create livable urban environments has become one of the priorities for policymakers in recent years. The introduction of innovative systems increasingly being introduced in modern cities, such as bike sharing, can certainly contribute to the spread of cycling and thus allow a radical change in the mobility habits of their citizens. This can be especially true for high-school students who are often otherwise accompanied by their parents in cars. This article aims to assess the influence that a bike sharing program for students, teachers and school staff has on modal share and on city mobility. As a case study, the city of Palermo was chosen, where the use of the car for home-school trips is prevalent. The “Go2School” project, which involves the creation of a bike sharing program for four schools, with the construction of cycle docks and cycle paths in the nearby areas, will soon become a reality. Thanks to appropriate surveys and questionnaires, a multinomial logit model was calibrated to estimate the modal share towards bike sharing for the students and evaluate the demand for this transport mode.

ABSTRACT. Car-Sharing Systems (CSSs) are becoming increasingly popular in urban areas replacing car ownership, especially in high-density cities. The most attractive CSSs give users the opportunity to make one-way trips. This behaviour, however, creates an unbalanced status between stations. For this reason, some users could leave the system because they may not find a car/parking place available near their origin/destination. As an alternative and more sustainable carbon-free mode of transportation, some CSSs are employing Electric Vehicles (EVs). A recent technology applied to EVs, called Vehicle-to-Grid (V2G), has allowed to sell energy, transferring it from vehicle batteries to an electric grid. In this paper, we propose to adopt Electric Cars (ECs) with V2G for one-way station-based CSSs. In particular, we suggest ECs distributions among stations at the beginning of each day, simultaneously making the most of V2G technology and satisfying CSSs customers’ requests. These distributions represent the final configurations that should be obtained through overnight vehicle relocations. The proposed model has been applied to a real-based test case achieving promising results.

ABSTRACT. In recent years, several ridesharing operators have launched their services across the globe. For these services, mobility requests arrive dynamically and have to be realized with a limited number of vehicles. The problem of request acceptance and route planning can be modeled as Dynamic Dial-a-Ride Problem (DDRP). Due to the limited transportation capacity of the shared vehicles, an important objective of the new service operators is to maximize the number of accepted requests. Since not all requests can be fulfilled, it is necessary to inform passengers immediately about the acceptance or rejection of their request. One way to achieve this is via feasibility check of the DDRP, which in this case must be performed within a very short computing time. The aim of this contribution is to examine the trade-off between computing time and solution quality as well as the effects of rescheduling during the feasibility check under realistic conditions of a typical urban on-demand ridesharing service. For this purpose, a Large Multiple-Neighborhood Search is proposed as an efficient approach to solve the DDRP. The analysis of different computing time limitation’s as well as the performance evaluation of the developed heuristic is based on computational simulation.

ABSTRACT. Two case studies that offered a shared ride-hailing service to work-commuters and students as an alternative to the private car and the public transport are analyzed and compared. The first case study took place during one week in Barcelona with the participation of 55 volunteers that used the service to commute from the city center to the most western district of the city. The second case study was based on the service test of MOIA in Hanover and involved the participation of 529 users that used MOIA to commute within Hanover and also as a first and last mile service. Results prove the suitability of this new mobility service for the use case commuting as well as for leisure trips and indicate the most important design factors to attract users and gain customer loyalty.

ABSTRACT. Endorsing transit effectiveness has become a key concern in promoting mobilities within urban and suburban surroundings. Many aspects in the burden of commuting have attracted increasing attention from researchers, and many interesting studies on this topic have been published already. This paper aims to contribute fresh evidence by examining the materials deriving from a thorough literature research for the sake of some of the case studies included in the European project SMART-COMMUTING. The research will include a combination of quantitative and qualitative approaches, aiming at examining which of the factors considered may be the best predictors of the efficiency of a given transit commuting policy for concrete applications on the project areas.

ABSTRACT. When expanding an existing public transport system a common practice consists of defining a set of candidate new lines and from them, choose those that will revert into a greater increase of the system's performance while keeping at a moderate level the associated amount of economical costs. In this work, a mathematical programming model for the selection of candidate lines in extensions of the public transport system is proposed based on the linearization of the model presented by Codina et al. (2013) and its application to situations of medium congestion. The model assumes that the pool of candidate lines is an input externally determined by technical criteria and it is focused in the compatibility constraints that the final set of selected lines must verify. These constraints include the overall throughput capacity of the stops/stations of the model, the availability of space for users to wait at stations, the resulting line capacities for passenger flows at line segments and the waiting times of passengers at stops. Solution procedures are based on the subgradient and Cutting Plane's algorithms and several small to medium size instances are successfully solved in the computational tests

ABSTRACT. School commuting is critical for modern societies considering its potential and long-lasting impacts in travel behavior of younger generations, today and in the future. Besides positive health impacts, exposing younger cohorts to more sustainable modes (e.g., walking, cycling and transit) is believed to be crucial to form future adults with more sustainable mobility decisions. Despite the vast research on school commuting, scrutinizing the foremost factors that determine the modal choice of households when children and youth commute to school is still challenging. Here, we carry out an analysis of the main factors that impact the willingness to shift to public transportation for school commuting, in the Lisbon Metropolitan Area. Also, we analyze the potential of hard and soft measures to change the households’ perception towards transit and their willingness to shift away from private car. While hard measures relate to interventions in the transport operation characteristics, soft measures act on the users’ side and aim to influence their collective and environmental consciousness. The preliminary results (from a sample of 1640 households) suggest that in order to achieve a modal shift towards public transportation, we should focus on improving transit services supply as the baseline for the whole system change.

ABSTRACT. This paper presents a comparative analysis of demand-responsive and fixed-schedule, fixed route operations for a simplified station-based feeder to mass transit scenario. Traffic dynamics, demand-responsive fleet coordination, and the behaviour of individual transit users are represented using a public transit simulation framework. Each operational strategy is simulated for varying levels of demand and two fleet compositions with respect to vehicle capacities and fleet size are compared. The services are evaluated based on resulting passenger waiting times, in-vehicles times and additional waiting time if one is denied boarding a fully occupied vehicle. Results indicate that dividing planned service capacity into larger fleets of smaller vehicles can provide a higher level-of-service to passengers. On an aggregate level, utilizing a fixed operational policy results in shorter and more reliable waiting times for levels of demand where there is slack in service capacity. In scenarios where planned service capacity is sometimes exceeded, the on-demand service provides a more even spatial distribution of passenger waiting times, relative to a fixed service.

ABSTRACT. The analysis of the functionality of road network and the quantification of travel times along the paths that vehicles follow is fundamental in the traffic engineering field. To analyze and estimate the flows that take to place in a network, some methodologies and tools have been used in the last decades. This paper revises the advantages of the use of data from plate scanning for the modelization and estimation of network traffic flows, considering also the definition of Origin-Destination matrix and traffic assignment into the network, as well as the quantification of travel times. The existing models and methods provide tools to determine the number of links which are valid to locate a device that collect data through the Automatic Vehicle Identification. This devices location can even lead a full observability of traffic flows, and a good estimation of travel times, since this techniques provide us more information than other known technique. For a correct application of this methodology in the real world, some challenges have to be faced as for example: it is necessary develop a device that combine efficiency and low-cost

ABSTRACT. Traffic Police enforcement is one of the main operations (along with engineering and education) to reduce road accidents by changing driver behavior. The traffic police is also responsible for handling events on the road, investigating road accidents, and prosecution. All these operations are based on three stages: (1) collecting data – from sensors embedded in the road, pictures and videos, cellphones, in-vehicle-data records, crowd sourcing and professional; (2) analyzing the data using a business information (BI) system and geographic information system (GIS); (3) Developing specific models for resource operations. The results of these operations are a reduction in the traffic offences and road accidents by severity in 2017-2018.

ABSTRACT. Data recorded by Automated Number Plate Recognition (ANPR) cameras can be used to determine several important traffic characteristics, such as real time travel time, travel time statistics, travel time reliability and OD matrices. In this paper ANPR data collected in Chinese city Changsha have been validated. Travel time extracted from ANPR data includes some outliers which are often caused by drivers who have an intermediate stop between two observation points or deviate from the straight route. Exceptional travel time reduces the validity of the estimation of the travel time and reliability. Firstly, the Rapid-Moving Window method is introduced to identify outliers. Afterwards, another method based on wavelet analysis is put forward to identify and remove the outliers in the travel time series. The wavelet analysis method is compared with the Rapid-Moving Window method and shows to be more accurate in outlier identification. The method for eliminating outliers in travel time can be implemented in real time to enhance the data quality for traffic network monitoring and management. After removal of outliers, the resulting travel times are used for the analysis of the relation between average travel time and the standard deviation and skewness.

ABSTRACT. Cities today must address the challenge of sustainable mobility, and traffic state forecasting plays a key role in mitigating traffic congestion in urban areas. For example, predicting path travel time is a crucial issue in navigation and route planning applications. Furthermore, the pervasive penetration of information and communication technologies makes floating car data an important source of real time data for intelligent transportation system applications. This paper deals with the problem of forecasting urban traffic when floating car data is available. A comparison of four deep learning methods is presented to demonstrate the capabilities of the neural network approaches (recurrent and/or convolutional) in solving the traffic forecasting problem in an urban context. Different tests are proposed in order to not only evaluate the developed deep learning models, but also to analyze how the penetration rates of floating cars affect forecasting accuracy. The presented experiments were designed according to a microscopic traffic simulation approach in order to emulate floating car data fleets, which provide vehicle position and speed, and to validate the obtained results. Finally, some conclusions and further research are presented.

ABSTRACT. Urban traffic currently affects the quality of life in cities and metropolitan areas as the problem becomes ever more aggravated by parking issues: congestion increases due to individuals looking for slots to park their vehicles. An Internet of Things approach allows drivers to know the state of the parking system in real time through wireless networks of sensor devices. This work focuses on studying the data generated by parking systems in order to develop predictive models that generate forecasted information. This can be useful in improving the management of parking areas. This research begins by looking at the state of the art in predictive methods based on machine learning for time series. This paper then introduces the recurrent neural network methods that were used in this research, namely Long Short-Term Memory and Gated Recurrent Unit, as well as the models developed according to real scenarios in different cities. In order to improve the quality of the models, exogenous variables like hourly weather and calendar effects are taken into account, and the baseline models are compared to the models that used this information. Finally, the preliminary encouraging results are described, followed by suggestions for corresponding future work.

ABSTRACT. Traffic conflicts based surrogate safety indicators have been applied extensively on real trajectories and in simulation. Such indicators can be useful to assess the safety of a given scenario without the need to use real crash data (which in many cases may be unavailable). Unfortunately, all traffic conflict indicators that are commonly used have a structural limitation: they are not able to consider potential conflicts with roadside obstacles or barriers and conflicts between vehicles which are travelling on non-conflicting trajectories. This limitation is a serious limitation since crash data analysis shows that at least 40% of fatal crashes are originated by single vehicle accidents against a fixed object or by vehicles travelling in opposite directions. The method of perturbation of trajectories presented in this paper allows researchers to implement new conflict indicators that can overcome the above indicated limitations.

ABSTRACT. Previous research has shown that people tend to drive slower under rain or snow conditions. Precipitation information was utilized to analyze its impact on traffic or to improve the traffic speed prediction. Conversely, traffic speed variation patterns of multiple roads may also provide indirect indication on weather conditions. In this paper, we analyzed an eight-month traffic speed dataset of New York City. With a seasonal trend decomposition model, residuals between the observations and the model were used as features to represent the level of anomaly as compared to the normal traffic situation. Based on the timestamps of weather records on sunny days and rainy/snowy days, such features were extracted from traffic data and assigned to the corresponding labels. A binary classifier was trained on six-month training data and achieved an accuracy of 91.74% when tested on the remaining two-month test data. We proved that there is a significant correlation between the precipitation events and traffic speed variation patterns, which can be used to learn a binary indicator. This indicator can be used to identify the anomalies of traffic infrastructures caused by precipitation events, which could improve the emergency response of cities where massive real-time traffic speed observations are available.

ABSTRACT. The paper presents a framework to estimate energy consumption of Electric Vehicles (EVs) by combining: (a) models derived from traffic flow theory and from mechanics of locomotion and (b) Floating Cara Data (FCD) from available Information and Communications Technology (ICT) devices. Existing energy consumption models may be classified into aggregate vs. disaggregate, according to the level of aggregation of variables related to driver, vehicle, and infrastructure. The proposed models have a hybrid nature: the aggregate component allows to estimating the values of vehicular speed and acceleration on a road link; the disaggregate one allows to estimating the discrete variability of EVs’ energy consumption inside a spatial-temporal domain. The energy consumption models are estimated using traffic data extracted from FCD. The proposed framework is structured into four steps: FCD processing, estimation of vehicular speeds and accelerations, estimation of resistance/energy consumption. The framework is applied in a pilot study area, composed by the backward (sub-)urban area of the port of “Porto delle Grazie” of Roccella Jonica (South of Italy). The preliminary results show that the methodology allows relative inexpensive and accurate calculation of EVs’ energy consumption and that it can be integrated into Intelligent Transportation System (ITS) applications.

ABSTRACT. The Jakarta greater area has high proportion of commuters who commutes daily from the municipals i.e. Bogor, Depok, Tangerang and Bekasi (JABODETABEK). Jakarta introduces the first Light Rail Transit (LRT), it serves the eastern part of Jakarta and Dukuh Atas (Jakarta-CBD). In fact, most of the commuter prefers either uses the private vehicles (i.e. car, and motorcycle) rather than public transport. Motocycle has higher accident risk compared to the traveling modes. A travel diary survey in this study captures the commuting pattern (prospective LRT passenger) which assists the stakeholder in optimising the network. This study identifies that the availability of a good and reliable schedule, and pedestrian facility encourage the commuter shifting to involve in active transport. 42% of the respondents accept to walk up 300m rather than uses a moto-taxi for their first and last-mile, if they have well-designed pedestrian network and facility. Public transport integration is a need for the commuter (17%), where they can transfer from one mode to the others. Pricing policy may attract the commuters, 69% of the respondents prefers zoning system rather than flat tariff system. Tariff differentiation (concession, off-peak ticket) may shift the commuter travelling pattern which assists public transport optimization and efficiency.

ABSTRACT. The main objective of this work is to quantify the energy consumption, travel-time, difficulty of each route and also safety levels for cyclists over different routes. Cyclists ride a conventional bicycle equipped with a GNSS to quantify energy required based on Bicycle Specific Power Methodology. Cyclists also wore equipment to record the heart rate called Vital Jacket and a video camera to record road conflicts between cyclists and cars. The aforementioned methodology was applied to three different routes chosen in the Portuguese cities of Aveiro (flat terrain) and Porto (hilly). For the flat city, the average energy expenditure was 44,5 Wh/km while for the hilly area the energy expenditure was 96,05 Wh/km. For each origin-destination pair by choosing an appropriate route it is possible to save about 28% energy in Aveiro and 35% in Porto. Regarding comfort, the average number of car overtaking manoeuvres to the bicycle was used as an indicator, while road safety was based on historical data. The tradeoffs identified and variation magnitude of variables analyzed suggest the information provided would be useful for cyclists with heterogeneous profiles as well as to support management authorities in order to maximize the attractiveness of the various solutions.

ABSTRACT. Managing automated vehicles in mixed traffic scenario necessitates special attention when introducing them into the market. In this research, we propose a strategy that operates connected and automated vehicles (CAV) to drive in a platoon which is led by a human driver. Following the strategy, detailed approaches are designed with the objective of increasing the time duration for which the vehicles drive automatically without increasing the total travel time. Reliable communication is essential for the application of these management approaches. In the examined scenario, all CAVs are assumed to be V2X enabled. CAVs will start communicating with each other and the roadside unit (RSU) when they enter the AV zone. The RSU will require CAVs to follow a certain platooning approach and CAVs cooperate with each other to form platoons. Micro-simulation is used for evaluating these approaches, as well as verifies it for communication reliability.

ABSTRACT. The hierarchy of urban transport networks constitutes the cornerstone of the traffic organization of a city. Until now, the conventional hierarchy approach has given exclusive priority to car movement, resulting in dysfunctional urban systems with unattractive public spaces and environmental issues. The need for a new hierarchy formulation that promotes sustainable means of transport is now more imperative than ever.

The aim of this paper is to create a methodological framework for formulating the strategic road network of a city. This method classifies the primary roads into 6 categories based on their topology, their trip length, their urban characteristics, the existence of trunk bus lines or metropolitan cycling routes as well as their current classification. The study area of the research is the city of Athens in Greece. The implementation of the proposed method can benefit Athens in many ways, such as promotion of sustainable means of transport, enhancement of urban vitality, environmental protection, unification of the urban fabric, accessibility improvement etc.

The methodological framework described, constitutes a street-oriented planning tool that gives priority to public transportation, walking and cycling. It can be applied to other study areas as well, acting as a roadmap for strategic road network formulation.

ABSTRACT. The discrete network design problem (DNDP) is a well-studied bilevel optimization problem in transportation. The goal of the DNDP is to identify the optimal set of candidate links (or projects) to be added to the network while accounting for users' reaction as governed by a traffic equilibrium. Several approaches have been proposed to solve the DNDP exactly using single-level, mixed-integer programming reformulations, linear approximations of link delay functions, relaxations and decompositions. To date, the largest DNDP instances solved to optimality remain of small scale and existing algorithms are no match to solve realistic problem instances involving large numbers of candidate projects. In this work, we examine the literature on exact methodologies for the DNDP and attempt to categorize the main approaches employed. We introduce a new set of benchmarking instances for the DNDP and implement three solution methods to compare computational performance and outline potential directions for improvement. For reproducibility purposes and to promote further research on this challenging bilevel optimization problem, all implementation codes and instance data are provided in a publicly available repository.

ABSTRACT. The aim of this paper is to present the methodology of solving complex multiple criteria problems with the presence of stochastic information. The author proposes a universal approach of ranking variants applied to solve decision problem in transportation. Since the nature of transportation process is stochastic, the evaluation criteria are expressed by nondeterministic values. Moreover, the preferential information is usually contradictory, with some level of uncertainty. As a result, the proposed methodology is composed of two phases, including the selection of the most suitable multiple criteria decision aiding (MCDA) method and its application using stochastic information. The first phase includes: suitability of MCDA method to both decision problem and decision maker’s preferences expressed during the modeling process, within the aggregation procedure and final results. The second phase is composed of four steps, such as: transformation of stochastic information into deterministic values, computation of relations between variants resulting in many deterministic rankings, calculation of stochastic relations between variants, and construction of final ranking. The proposed methodology is verified on the exemplary decision problem in transportation. The application of the procedure increases the credibility of the final decision, giving an opportunity to use deterministic MCDA method in the stochastic transportation process.

ABSTRACT. Bike-sharing systems operate in a number of cities around the world, aiming to promote sustainable urban mobility. Successful management of these systems is to a large extent linked to the optimal distribution of bicycles, which implies the accurate prediction of demand for rentals and returns at each station within the day. For this purpose, a tool for predicting bike demand for rentals and returns and visualizing the results has been developed and is presented in the present paper. Different predictive models, based on machine learning regression algorithms are trained and evaluated. The tool is tested using data from the bike-sharing system that operates in the city of Thessaloniki, Greece for which the results indicate that the tested system’s utilization is highly correlated to the location and spatial characteristics of a station, as well as the season of the year and time of day. The proposed machine learning algorithms use custom engineered features to learn those correlations and achieve the highest possible performance.

ABSTRACT. Although there has been an increase in the number of researches involving humanitarian logistics, few studies address slow disasters, such as drought. Although this phenomenon is one of the worst types of disaster in terms of casualties, water distribution to those affected by the drought still lack academic development, and few studies guide local public managers regarding the efficient use of available resources, guaranteeing a high service level to the population. This paper presents a procedure for the implementation of transport and routing of water delivery. By this procedure, we can evaluate the strategy of complementary use of the water supply of drilled artesian wells. It also provides a contextualized routing tool for the practices and policies of motorized water distribution, commonly found in scenarios of water shortages. The algorithms have been implemented considering the capacitated vehicle routing problem with time windows (CVRPTW), providing a friendly-user tool that helps choosing water sources to be activated and in the definition of the routes between these and the points supply. Results have been validated in the Brazilian Semi - arid region, showing improvement in the current logistical performance of drought coping, in advantageous scales of operation in terms of costs and efficiency.

ABSTRACT. The paper presents a Multiple Criteria Analysis (MCA) of trams, based on customers’ specifications and open bid regulations in different countries, such as: Czech Republic, France, Germany, Hungary, Poland and Turkey. We propose a universal assessment procedure for trams to be used in urban, public transportation system, based on the principles of Multiple Criteria Decision Making/ Aiding (MCDM/A). The analysis of trams is formulated as a multiple criteria ranking problem. The proposed paradigm includes: the definition of the group decision maker – DM and stakeholders, selection of variants (trams), formulation of the consistent family of criteria, recognition and modeling of the DM and stakeholders’ interests/ preferences, performance of the computational experiments, generation of the ranking of trams. We perform a series of computational experiments with an application of selected multiple criteria ranking methods (Electre III/IV, AHP, ANP, Promethee). The generated rankings result in a comprehensive analysis of the quality of trams in different countries and the suitability of selected MCDM/A methods for the evaluation of trams.

ABSTRACT. The European Commission proposed emission reduction targets on car manufacturers by 2030, to reduce CO2 emissions. The present analysis aims to assess whether the implementation of policy bears positive or negative impacts across different household income classes. The analysis is based on a quantitative post-processing analysis of PRIMES-TREMOVE and GEM-E3 scenarios. The scenarios used in this paper are the ones that supported the quantitative analysis of the EC proposal. The paper draws policy information by observing at each case, if the fuel savings of the vehicles marketed under the more ambitious targets on cars are enough to outweigh the higher purchasing prices. At the end of the exercise, a sensitivity analysis is conducted by changing the vehicles’ depreciation rate, the assumed economic lifetime and the household discount rates, to examine the degree of change and impact of those parameters on the answer to the policy question. The analysis shows that households that purchase more efficient vehicles in the second-hand car market benefit to a greater extent from the annual fuel savings by paying a fraction of the additional cost of the first owner. The sensitivity analysis shows that such conclusion is supported for a range of assumptions.

ABSTRACT. It is well-known that the transport sector activity causes different detrimental effects on the environment and the social welfare of our society. These negative impacts are usually presented as negative externalities which increase the logistic costs. In this paper, we solve a Sustainable Team orienteering problem (STOP) in which those negative impacts are mitigated by minimizing the travel times. It consists of maximizing the collected profit by a driving range, visiting each client at most once. Thus, if a customer is visited and the maximum driving range is completed, no profit is collected. In this work, the travel time varies according to the routing plan and the system state. A learnheuristic-based approach is proposed to maximize the collected profit and minimize the negative impacts of transport, under dynamic conditions. A series of computational experiments contribute to illustrate the advantages of the proposed method.

ABSTRACT. The last-mile is the most inefficient and most expensive part of a supply chain and comprises up to 55% of total delivery costs. A system of parcel lockers operated by a non-competing company, usable by any logistic service provider can contribute to make goods distribution and the last-mile more sustainable. The research project “SoWAS” investigates the development of such a system and will evaluate generated changes in costumer’s mobility behaviour by an online panel in a living lab area. This paper presents the approach of the online panel and its first application on two focus groups. Besides findings of approach testing, the collected data give interesting and precise insights into costumer’s mobility behaviour in connection with received and dispatched/returned parcels – a field with very poor available data. Based on the collected data we will estimate changes in mobility behaviour and reduction potential concerning vehicle kilometers, emissions and pick-up trips due to the introduction of a parcel locker system.

ABSTRACT. The sustainable development of urban areas, characterized by the expansion of university sites, can benefits from car sharing services, expanding the potential of public transport, especially when the population is almost represented by off-side students and commuters. Based on this premise, the main focus of this work is on the investigation of a set of attributes able to interpret and measure the propensity of students to join car sharing services. The paper provides a discussion about the potential implementation of a car sharing service, including a presentation of the state-of-art practices and particularly referring to the case of the Enna (Italy), where students constitute a significant percentage of residents. The propensity was assessed by a survey based on the Likert Scale and a chi-squared test of independence was calculated to check the difference between the expected and observed frequencies for the considered attributes and for several of their combinations. Students have been involved via a wide consultation survey and the analysis was conducted by using mixed Internet/paper survey data. Results are useful to understand their heterogeneous preferences and to pave the way for a well-thought-out design of a new shared transport service

ABSTRACT. Repositioning is the main set of operations on station-based bike-sharing systems. It is necessary to avoid shortages of bikes on certain stations and parking slots on others. Therefore, its cost optimization is key to determine system success or failure. Scientific literature has dealt many times with this problem, mostly modelling it as a routing problem and providing solution algorithms. Those models are considered “static” or “dynamic” if the repositioning takes place when system is closed or still open. This paper addresses those problems by redefining dynamic repositioning as an alarm-based set of operations which complements the routing-based solution and improves the overall performance.

ABSTRACT. In accordance with the spread of ridesharing services, users’ data, how, when, where and for what purpose they use the service, are gradually collected in the world. However, there are few studies that deal with empirical situations in Japan. In addition, ridesharing services in Japan are not yet popular compared with that in Western countries thus far. Therefore, in this study, we aim to show the present situation of ridesharing behavior in Japan. We conduct the empirical analysis by using actual long-distance peer-to-peer ridesharing data in Japan. Firstly, we classify ridesharing drives into three classes by OD pairs. Next, we formulate the binomial probit model that explains matching success for each drive class. Estimated model shows that departure time and day, days to departure from registration date, page views and driver's past experiences are important factors for successful match make.

ABSTRACT. In the last years, bus passengers require reliable and real-time information but the real-time information systems developed still do not fully satisfy their needs. Within the framework of the project HARMONY, a mobile transit app has been developed in Madrid Region to improve the information that metropolitan bus passengers receive from operators, and enabling an option for passengers to send incidences to operators in real-time. A SWOT matrix has been built and a two-stage consultation to passengers has been carried out. This allows to know the current transit app market and to detect gaps in the users’ needs, leading to the features to be implemented. None of the existing apps allow a bi-directional communication between operators and passengers. Survey results reveal that apps like Google Maps do not compete with specific transit apps that include official, real-time information because of the type of data offered.

ABSTRACT. Digitalization enables travelers to access an ever-increasing number of mobility services. Creating door-to-door multimodal travel itineraries from individual mobility services is challenging, since the impact of individual travelers’ preferences on complex multimodal travel itineraries is not transparent. To support travelers in their individual choice of relevant itineraries, travel websites create variants of multimodal travel itineraries using multimodal shortest path approaches. Then, they let travelers specify their individual preferences further to reduce or extend the set of travel itineraries. However, due to the complexity of the underlying search space, it is challenging to individualize the created set of travel itineraries systematically. In this paper, we use solution sampling to create request-specific meta-information about the complex solution space of multimodal travel itineraries. We identify intervals of and relationships between travel parameters such as travel time, price, number of transfers and mode choice by applying multimodal shortest path approaches with solution sampling. The derived request-specific meta-information is used to individualize a traveler’s set of multimodal travel itineraries. We examine solution sampling in a laboratory setting with data on scheduled and non-scheduled transportation services from German transit networks.sampling in a laboratory setting with data on scheduled and non-scheduled transportation services from German transit networks.

ABSTRACT. In recent times, new models of commerce have incorporated new decisions and constraints which have led to new variants of classical problems in supply chain management. Modern advances in Information and Communication Technologies have increased the number of marketing channels that are available to consumers. This paper focusses on the new ``omnichannel'' delivery concept for the retailing industry which addresses the replenishment of a set of retail stores and on the direct shipment of the products to customers (last-mile delivery) within an integrated VRP formulation. The VRP in omnichannel distribution systems consists of a group of retail stores that must be served from a distribution center and a set of online consumers that must be served by the same fleet of cargo vehicles from these retail stores. Since the VRP in omnichannel distribution systems is an NP-Hard problem, we propose a savings-based heuristic for solving large-size instances the VRP in omnichannel retailing. Results show that the proposed heuristic is able to find feasible and competitive solutions in a very short computational time.

ABSTRACT. Tourism is one of the largest growing industries worldwide, and so are the increased safety concerns. This is due to increasingly frequent and severe natural hazards as well as terrorism, where large crowds of tourists can be targeted. Furthermore, tourists are often less informed and are therefore more vulnerable to be trapped in chaotic situations. In such situations, normally we are interested in fast evacuation routes. However, tourists, especially during emergency situations, are prone to orientation and navigation errors. Such errors can be avoided by providing electronic guidance at various intersections along the evacuation path (controlled intersections). This study formulates the above-mentioned situation as the shortest path problem with stochastic routing. The stochastic routing is the probabilistic selection of an outgoing arc at each node. As it is practically difficult to equip every intersection with guidance devices, a multi-objective model is developed. The model simultaneously minimizes the number of controlled intersections and minimizes the gap between the actual evacuation route and the optimal evacuation route. The problem is formulated as a stochastic multi-objective problem. A Pareto-front of solutions is obtained using a genetic algorithm and a simulation.

ABSTRACT. The concepts of unmanned aerial vehicles and self-driving vehicles are gaining relevance inside the smart city environment. This type of vehicles might use ultrareliable telecommunication systems, Internet-based technologies, and navigation satellite services to decide about the routes they must follow to efficiently accomplish their mission and reach their destinations in due time. When working in teams of vehicles, there is a need to coordinate their routing operations. When some unexpected events occur in the city (e.g., after a traffic accident, a natural disaster, or a terrorist attack), coordination among vehicles might need to be done in real-time. Accordingly, this paper analyzes how the use of extremely fast, yet effective, biased-randomized heuristic allows for ‘agile’ optimization of routing plans for drones and other autonomous vehicles.

ABSTRACT. In this paper, a mathematical programming-based model is presented for evaluating the impact that proposals for infrastructure improvement and capacity expansion can have on a mixed railway network, mainly oriented to increase freight transportation. In this model, the authors have considered extensions of elements of an existent mixed railway network, jointly with actions on the network with relative smaller cost, such as the inclusion of new sidings or new gauges in several rail segments, expansion of classification terminals or stations, and also capacity enhancements by adding new blocking/control systems at specific locations. These aspects are usually not taken into account in models for regional planning. Our approach, rather than a model of railway capacity expansion can be considered a mixture of capacity-expansion with network design. The model is tested on a small regional network of the Mediterranean Corridor, and the computational results show its applicability to larger networks.

ABSTRACT. While cooperative services have been almost fully deployed in the road sector and are already being implemented in various cities in Europe as a pre-requisite for the introduction of autonomous vehicles, few attempts have been made in the same direction for the rail sector. This study proposes a system that aims to improve safety and minimize risk in the meeting point between road and rail, known as level crossings, by monitoring the location of floating road vehicles via a mobile device application. A neural network predictive model for estimating time of arrival of trains is also utilized. The safety system has been implemented and tested under real life conditions in the city of Thessaloniki, Greece.

ABSTRACT. This paper studies how time delays spread through the railway system. We evaluate timetable and rolling stock interaction in relation to medium size delays. In particular, we study the connection between arrival and departure delays for trains that are operated by the same rolling stock individual. We studied one year of data for trains circulating in the Oslo area. Research questions in this study are: • What is the connection between arrival and departure delays for rolling stock individuals? • How can delays be tracked through the time table and the rolling stock circulation plan?

Punctuality analysis was performed to investigate if there is a connection between arrival and departure delays. A correlation analysis of the arrival delays and the corresponding departures was done in order to find the possible connection. There was a connection between the size of arrival delays and the size of the corresponding departure delays. We found a threshold at around 20 minutes. An arrival delay over 20 minutes is expected to cause departure delays. Arrival delays of less than 20 minutes are to a larger extent absorbed by the production plan.

ABSTRACT. Blocking constitutes an important rail freight operation, by which cars with potentially different origins and destinations are grouped to be then moved and handled as a single unit, yielding important economies of scale. We address the block tactical planning problem for intermodal railroads, which has been little studied so far. We propose a new service network design model that explicitly considers the specificity of intermodal rail and which may be solved using commercial software for realistic sizes. We illustrate the performance and interest of the proposed method through an extensive case study of a major North American railroad

ABSTRACT. Nowadays, there are several companies providing processed or raw global positioning system (GPS) measurements from fleets of commercial vehicles, internet applications or automobile companies. The aim of this paper is to deepen the understanding on the GPS data applicability for transportation modelling purposes by providing systematic and quantified insights into collected data representativeness in describing individuals’ route choice. Unfortunately, real data often contain noise, uncertainty, errors, redundancies or even irrelevant information. Useless models will be obtained when built over incorrect or incomplete data. This is why pre-processing is one of the most critical steps of data analysis in any of its forms. However, pre-processing has not been properly systematized yet, this paper focuses on pre-processing steps required in GPS tracking data together with a proposal to systematize it. Furthermore, the aggregation level is at waypoint location for low latency GPS positions in the trajectory in their trip. Travel time reliability in OD paths is addressed among other OD path characteristics. Route choice in dense urban networks face multiple possibilities and data from new technologies offers the opportunity to understand route choice behavior.

ABSTRACT. Nowadays, road safety is an issue particularly relevant because of the increasing occurrence of road accidents. For this reason, it is very important to analyse accident severity and the factors influencing it. This paper aims to investigate on the characteristics that can influence the severity of an accident, grouped in different categories, related to road, external environment, and driver. The proposed analysis has the specific objective to discover the possible differences between the various types of accident (e.g. clash, collision, pedestrian investments) in terms of factors influencing the severity expressed through information such as the number of dead or injured persons. The method proposed to this aim is represented by mathematical models able to determine the weight of each factor involved in the analysis on accident severity. The preliminary results suggest that there are interesting differences between the types of accident.

ABSTRACT. Road pricing basic rationale argues that congested occur as the result of increasing number traveller which used private transportation mode, instead of public transport. Road pricing could influence road users’ choice for transportation method. An electronic road pricing (ERP) is planned to be implemented in 2019 in Jakarta, the capital of Indonesia, while the public acceptance of the policy is not yet be discussed. This study aims to describe public acceptance of road pricing with the influence of the differences in socio-demographic condition (Model 1), traffic management strategies perception (Model 2), and road pricing revenue allocation perception (Model 3). The data is collected from 356 respondents in Jakarta. This study uses logistic regression (logit) model to facilitate discrete nature of the data especially public acceptance as a dependent variable that is categorized as an accept/not accept answer. Preliminary results show that Model 1 is not significant; while Model 2 shows that perception on road pricing is significant; and finally Model 3 is significant especially if road pricing revenue is allocated to develop public transport, to increase road connectivity, and to protect the environment.

ABSTRACT. Compared to the other public transport systems, air transport has received limited attention on the assessment of service quality. This paper aims to explore the factors employed to assess the airport service quality, taking as case study the International Airport of Lamezia Terme, a peripheral airport placed in the south of Italy. Specifically, through a Principal Component Analysis (PCA) latent factors influencing service quality were identified and the dimensionality of the phenomenon was reduced. After that, a Structural Equation Modelling (SEM) approach was performed in order to define the relationships among the latent variables, and between the observed variables and the latent ones. For these purposes, we analyzed a database derived from Customer Satisfaction Surveys (CSS) conducted during 2015-2016 in the Lamezia Terme airport. The results confirm that overall airport service quality is significantly related to latent factors such as the accessibility to the services, the control operations in the terminal, and the comfort offered to the passengers.