ABSTRACT. Fleet tracking technology collects real-time information about geolocation of vehicles as well as driving-related data. This information is typically used for location monitoring as well as for analysis of routes, vehicles and drivers. From an operational point of view, the geolocation simply identifies the state of a vehicle in terms of positioning and navigation. From a management point of view, the geolocation may be used to infer the state of a vehicle in terms of process (e.g., driving, refueling, maintenance, or lunch break). Meaningful information may be extracted from these inferred states using process mining. An innovative methodology for inferring process states from geolocation data is proposed in this paper. Also, it is presented the potential of applying process mining techniques on geolocation data for process discovery.

ABSTRACT. This paper presents a novel approach for exploring vehicle daily mobility patterns using Floating Car Data in the Paris region. The objective is to reveal vehicle types and analyze usage patterns on different roadway sub-classes. Firstly, mobility representative features are recovered to recognize vehicle activity context, which includes modeling single trip pattern in terms of time windows, traveling distances, and speed; and building vehicle mobility profile of trip combinations. Based on that, a two-step clustering algorithm is developed to explore the constitution of trip patterns and cluster vehicle types. Characterization analysis is conducted to find out outstanding features of clustered groups, thus helping to categorize the vehicles behavioral types. As a result, 4 major vehicle types were identified over the Paris region with the 2 comparative leading groups as those mainly composed by morning-activity trips and long-distance trips. Then, statistical association assessment by Configural Frequency Analysis is employed to examine the usage intensity on different roadway classes of identified types. The association analysis reveals that the identified types have statistically significant differences in the usage of different roadway classes. Furthermore, this approach can be expected to provide more representative results with more generalized sampling by the development of connected vehicles.

ABSTRACT. This work investigates network-related trajectory features to unravel trips that the most contribute to the system under-performance. When such trips are identified, features analysis also permits to identify the best alternatives in terms of routes to make the system to its optimum. First, descriptive analysis is carried out on trajectories obtained from reference dynamic traffic assignment (DTA) simulations in a real-world network, based on User-Equilibrium (UE) and System-Optimum (SO). This analysis helps us (i) to target the trajectories to be changed, and (ii) to identify their main features (path marginal costs, network-related features such as betweenness centrality and traffic light parameters, etc.). Then, we give targeted users predefined paths and re-run DTA simulation under UE condition, in order to evaluate the efficiency of targeting methods. Results show that changing routes of 10% users whose trajectories are (i) with high betweenness centrality or (ii) with small mean MFD capacity, can already bring more than 6% of total travel time (TTT) reduction with respect to TTT in UE-simulation, while the TTT reduction from UE to SO simulation is about 14%.

ABSTRACT. In recent years, studies on network vulnerability have grown rapidly in the fields of transportation and complex networks. Even though these two fields seem to be closely related, it is not clear how these two fields have evolved over time and how the two fields have influenced each other. In this study, in order to add clarity comprehensively and objectively, we analyze a citation network where nodes represent publications and directed links represent citation relations. We collect publication records from the Web of Science and construct a citation network with 1,181 nodes and 4,601 links. Then we analyze the giant weakly connected component with 705 nodes and 4,584 links. By performing the community detection algorithm, we identify seven communities representing different research domains: 1) transport vulnerability, 2) metro and shipping, 3) resilience, 4) topological vulnerability, 5) cascading failure, 6) interdependent networks, and 7) scattered community. Then we identify the major research development in each research domain with main path analysis. Finally we reveal the citation patterns among different research domains.

ABSTRACT. The object of the paper is to study issues related to reactive dynamic assignment in the context of communication, and specifically to analyze some impacts of information on reactive DTA (dynamic traffic assignment). New means of communication and new services will make instantaneous information on the network traffic state available to drivers / vehicles. Instantaneous and long range interaction between all drivers /vehicles will become feasable, with possible negative effects on the network traffic dynamics. In order to analyze such effects, the paper proposes the following: i) a dynamic multimodal macroscopic traffic flow model, for analysis an also prediction, based on the GSOM (generic second order model) approach; ii) a methodology for travel times estimation, compatible with the GSOM model; iii) a simple test setting for analyzing the impact of information on DTA; iv) and a simple control strategy based on a dynamical systems approach, designed to foster user optimality. Preliminary results show that instantaneous travel time constitutes an information liable to induce unfavorable impacts and that predictive travel times or the simple control strategies yield better traffic dynamics.

ABSTRACT. The paper presents a methodology to measure the dynamic resiliency of a public transport network and its application to New Delhi bus network. Variability in service supply that arises from normal day-to-day service fluctuations leads to negative passenger experiences and therefore satisfaction. The paper explores the utilization of dynamic data to develop a systematic approach analysing disruption impacts with respect to service-wide capacity changes due to fleet failures. In order to measure the dynamic resilience, an index is formulated that covers the magnitude of changes in the network performances in terms of robustness, reliability, and residual capacity; measures that describe the network complexity in a physical and a service operational specifications. In this regard, the relationship between system components is characterized by network topology, passenger flow, and service performance dynamics. Our proposed approach is able to capture the dynamics of the resilient system and to quantify the extent to which operational strategies are capable of delivering efficient systems. We illustrate our model by an application to the New Delhi bus transport system and demonstrate how system resiliency for different fleet-failure states can be quantified.

ABSTRACT. In a road network, drivers typically seek to minimize their own travel time, often affecting system-wide performance. For an efficient traffic assignment (TA), besides concerns with travel times, traffic managers should not neglect system-wide level of pollutant emissions. Measuring road traffic emissions can be costly and models based on vehicle-specific parameters with many input variables have been suggested in the literature. This paper proposes a simple way to quantifying carbon dioxide (CO2) and nitrogen oxides (NOX) emissions with only average speed as input variable and presents a multi-objective TA approach that seeks to minimize system-wide travel time, distance travelled (associated with fuel consumption) and global and local pollutant emissions. A real-world case study on an intercity corridor with many alternative routes between two zones is presented. Experiments considering TA based on travel time, and on time, distance travelled and pollutant emissions are reported. Results highlight that system optimal distribution based on the suggested multi-objective TA based on three components yields savings in terms of distance travelled (2.6%) and emissions (1.3% for CO2 and 1.1% for NOX), but penalizes travel time 3% translated in an increase of 20sec per vehicle, when compared to the solution only focused on minimizing travel time.

ABSTRACT. A trial-and-error congestion pricing scheme finds the optimal toll based only on observable information (e.g., travel time) without information on travelers’ personal preferences that are often unobservable in practice (e.g., value of time, scheduling preference). This paper proposes a trial-and-error pricing scheme for the morning commute problem, also known as the departure time choice problem and Vickrey's bottleneck model, with day-to-day dynamics. We mathematically prove that the proposed scheme is guaranteed to find the social optimal congestion toll under somewhat restrictive assumptions. Then, we numerically confirm that the proposed scheme converges to the social optimal toll fairly quickly.

ABSTRACT. It is proposed an integrated framework for timetable design which combines the requirements of service planners, which usually adopt a purely macroscopic point of view of the demand requirements and the conflicts that these timetables may arise at stations. These aspects may not be considered by the service planners since a microscopic analysis of the stations structure is necessary considering possible conflicts at platforms. The proposed integrated model guarantees feasible timetables balancing performance indexes for the passenger and operator interests while guaranteeing feasibility. The differences between optimal timetables for the service planners and those who are operational at stations are kept within specified bounds. The resulting problem has been modeled as a multi-objective Mixed Integer Linear Programming problem. The model can be solved to optimality in reasonable computing time as shown in the tests using the Madrid-Zaragoza-Barcelona high speed railway network. For this case a Pareto frontier analysis is performed showing that the problem is well-posed. Also, based on the previous model a heuristic procedure is developed in order to obtain timetables which avoid as much as possible concentrations of arrivals/departures to/at stations, thus allowing for a better recoverability and robustness of the timetable in case of unexpected disruptions.

ABSTRACT. The paper deals with a problem of distribution network design with regularity of deliveries. A combinatorial model of two-tier distribution network is proposed, where the objective is a cost-oriented function, taking into account all transport and handling operations, excluding an inventory costs. The model is universal, and it constitutes a skeleton of decision making on network design. However, scenario-based extension of the optimisation model allows to reflect specific business factors as: depot’s relocations, regularity limits, time limits of deliveries, and vehicle’s capacity. Each scenario is a combination of those factors. A constructed model has been experimentally applied to optimise the distribution network of drugstore companies, which is one of the key player on the Polish market. The current configuration of its network is composed of 1 central warehouse, 15 transshipment depots and 256 drugstores. There are considered five scenarios of distribution network configuration. Each of them have been optimised and those results have ranged from nearly 6% increase to more than 30% reduction of objective function, referred to current state. A scenario adopted upon managerial decision has guaranteed over 12% reduction, and validation of the result and a proposed procedure has been proved in practice.

ABSTRACT. We examine an innovative system for organizing deliveries in a collaborative fashion for an n-tier hyperconnected city logistics system. We focus on the tactical planning of services within the first tier of the system, i.e., from external zones generally located on the outskirts of the city (logistics platforms, urban/city distribution/consolidation centers, etc.) to satellites from which goods are distributed to final customers, and introduce a new optimization model for that purpose. The key distinctive feature of this model is that we consider a coalition of carriers and logistic operators who share their resources (fleets of vehicles and warehousing capacity) and information flows to provide more effective services, thus lowering costs and environmental impact. Preliminary computational results confirm the attractiveness of the envisioned system.

ABSTRACT. The article presents the method of participatory multi-criteria analysis, which allows to include unorganized groups of stakeholders in the evaluation process. Special attention has been paid to transport users, who in urban conditions, can use many different means of transport. The procedures of selecting user groups, the aggregation of preference models within groups as well as the aggregation of group preferences were presented. The method was applied to identify stakeholder preferences with respect to the redevelopment options of Rondo Rataje in Poznań.

ABSTRACT. The introduction of autonomous vehicles will increase the vehicle heterogeneity in our roads. It is claimed that these vehicles will be able to achieve lower spacings for the same speed than human drivers. Therefore, a good understanding of the influence of heterogeneous driver behavior on traffic flow macroscopic characteristics is crutial. This paper presents a stochastic LWR model by introducing heterogeneous, i.e., vehicle dependent, jam densities. The model is solved in Lagrangian coordinates and the probabilistic nature of the model allows for investigating the impact of driver heterogeneity on macroscopic relations of traffic flow, both analytically and through simulations. The results show that macroscopic characteristics of the model are consistent with the deterministic version with an equivalent jam density, that is the harmonic mean of the distribution.

ABSTRACT. Autonomous vehicles (AV) will be present in freeways, and they will have to share current infrastructure with human driven vehicles. This mixed traffic scenario needs to be planed, as some research points that a mismanagement of AV could lead to a capacity decrease. This can be solved with AV platoons, which are strings of vehicles traveling at very short headways. Some research exist to that end, but all of them is done using microsimulation tools, which are very powerful but have the shortcoming of strongly rely on an uncertain calibration. The more robust macroscopic tools, have no models yet suitable for platoons. The research presented in this paper wants to fill the gap, by providing a generalized macroscopic model based on the LWR theory for platoons. This is achieved by considering a freeway with two types of lanes (platoon and general purpose) and two types of vehicles (AV and non AV) and establishing some rules of behavior. Among the rules, platoon formation has to be set, two options are presented: cooperative and opportunistic. Results show that given the appropriate conditions, great capacity increments can be achieved, multiplying by factors of more than two the current observed freeway capacities.

ABSTRACT. The future use of automated vehicles involves a great deal of uncertainty regarding various aspects, including the interaction with vulnerable road users. This interaction has been studied only to a limited extent so far. As a first step in examining this interaction, there is a need in detecting relevant conflict points involving both types of users (automated vehicles and vulnerable users). These conflict points should reflect both safety issues; concerning vulnerable road users and operational issues involving automated vehicles. We present a methodology for identifying such conflict points, in order to simplify the selection of adequate instances to be used in studies involving automated vehicles. The presented methodology is based on two pillars: the statistical analysis of conflict points for the identification of safety problems based on accident statistics, and the analysis of operational aspects concerning the use of automated vehicles. The proposed methodology, will be demonstrated on the city of Braunschweig.

ABSTRACT. This study provides insights into traveler-related attributes affecting the choice of future autonomous vehicles (AVs) and explores the importance of integrating those attributes into agent-based simulations and service optimization assessments. For this purpose, an online survey was carried out to collect data on travelers of the greater Paris region and their behavior regarding autonomous vehicles. In addition, this paper identifies AV taste variation among individuals as well as the criteria behind their willingness-to-use a shared autonomous vehicle service depending on their current mode of transport. The paper shows how traveler-related attributes are relevant to studying a shared autonomous mobility system and how they can enhance the accuracy of agent-based models and the traveler preference dimension in optimization models.

ABSTRACT. The proposed model of the bus lines synchronization in a transfer node is based on simulation of demand for changing the public transport lines with a genetic algorithm as a tool to obtain some rational solution. The described approach considers stochastic nature of demand for public transport and provides some rational solution of the synchronization problem. The total waiting time of passengers at the given transfer node is used as an objective function in the synchronization problem. Synchronization is implemented due to time shifts at the transport lines timetables for the transfer node; these time shifts are represented as chromosomes of a genetic algorithm. In order to evaluate the objective function, simulations of public transport lines servicing the passengers in a transfer node were provided. The developed mathematical model is implemented in Python within the frame of a class library for modeling of public transport processes. A case of the Wiślicka transfer node in a public transport system of Kraków city is applied in order to illustrate the developed synchronization procedure.

ABSTRACT. The paper refers to the development of transportation/ logistics solutions in developing and politicaly and militarly sensitive regions. It is focused on providing a rational paradigm of decision making that would allow the decision makers (DMs) to produce a desirable, efficient, and sustainable freight transportation/logistics solutions. We focus on the design and evaluation of a freight logistic system delivering goods (electrical cables) to Iraq. The intuitively designed transportation/ logistics solutions are assessed with the application of the Multiple Criteria Decision Making/ Aiding (MCDM/A) methodology. Our approach is based on thorough analysis of the decision situation (internal and external conditions; decision makers, stakeholders; specific local constraints; existing expectations and preferences). We take into account the trade- offs and constraints (economic, political, social) to reach a compromise and most desireable solution. We carry out a series of computational experiments with the applcation of selected MCDM/A ranking methods, e.g. ELECTRE III/IV and AHP. We develop a universal decision model and adapt it to specific local circumstances and characteristic features of the applied decision tools, i.e. ELECTRE III/IV and AHP methods.

ABSTRACT. In the immediate aftermath of a natural disaster, physical infrastructure, such as roads and bridges, are often destroyed and transport capacity is extremely limited or non-existent. As a result, access to affected areas becomes very difficult or even remote. In this scenario, helicopters are the most appropriate vehicles to reach the victims. However, the planning of air transport operations in the context of a disaster response has a high degree of complexity, so that operational research has significant application and potential contribution to the area. In this context, this paper proposes a procedure that aims to optimize the use of helicopters in response operations to small and medium-scale natural disasters. The proposed procedure seeks to minimize the total time of operation and the mobilization of air resources during last mile deliveries in relief operations. This procedure was applied on a real post-disaster scenario, taking as basis the characteristics of the response operation to the floods, occurred in 2011, in the mountain region of the state of Rio de Janeiro, Brazil. Results indicate the developed methodology as a viable tool to aid in the decision-making process regarding the use of helicopters for last mile deliveries in the humanitarian supply chain.

ABSTRACT. The article presents a comparison of methods of evaluation of establishing the rankings of variants on the example of the location of P&R lots (car parks) in Bydgoszcz (a medium-sized city in Poland). The methods to be compared were based on the procedure defined by municipal authorities in the public procurement (a simple evaluation based on a weighted average) on the one hand, and at the same time, on the well-known methodology of multi-criteria decision making. The assessment was carried out for 15 predefined locations, and both methods used the same set of criteria and sub-criteria. In order to determine the value of the criteria, among others, the travel model of the Bydgoszcz agglomeration built in the VISUM tool was used. As a result, it was possible to simulate the creation of P&R lots in selected locations. The preferences of the decision-maker (the municipal authorities) were expressed in a direct interview with their representatives. As a result of all the calculation work, two final rankings were obtained for each method, which allowed to draw final conclusions.

ABSTRACT. This paper aims to explore the performance of autonomous mobility-on-demand systems (AMoD) with the coordinated formation of vehicle platooning. In this study, an agent-based model (ABM) is developed to explicitly simulate the operations of both vehicle platooning formation and interactions between shared automated vehicles (SAVs) and real-time demand requests. The objective is to capture the stochastic behavior of SAVs as trip makers, and then assess the performance of the AMoD system with the mechanism of coordinated formation of platoons. We conclude that the impact of vehicle dispatching strategies in the AMoD system with vehicle platooning formation predominately affects the average waiting time and system capacity to transport travellers as a whole; however, vehicle platooning, to some extent, could lengthen the in-vehicle travel time of participating vehicles. The hold-on time (imposed delay) of leading vehicles in order to form a platoon could affect the average delay of vehicles part of those platoons. The developed ABM provides the first insight into the impact of the pervasive formation of vehicle platooning on the performance of the AMoD system.

ABSTRACT. The growth of urban areas has made taxi service become increasingly more popular due to its ubiquity and flexibility when compared with, more rigid, public transportation modes. However, in big cities taxi service is still unbalanced, resulting in inefficiencies such as long waiting times and excessive vacant trips. This paper presents an exploratory taxi fleet service analysis and compares two forecast models aimed at predicting the spatiotemporal variation of short-term taxi demand. For this paper, we used a large sample with more than 1 million trips between 2014 and 2017, representing roughly 10% of Lisbon’s fleet. We analyzed the spatiotemporal variation between pick-up and drop-off locations and how they are affected by weather conditions, irregular events and points of interest. More, based on historic data, we built two models to predict the demand, ARIMA and Artificial Neural Network (ANN), and evaluated and compared the performance of both models. This study not only allows the direct comparison of a linear statistical model with a machine learning one but also leads to a better comprehension of complex interactions surrounding different urban data sources using the taxi service as a probe to better understand urban mobility and its needs.

ABSTRACT. Under a likely transitional stage of co-existence of connected and automated vehicles (CAVs) and conventional vehicles (CV), this study explores the potential effects of CAVs to reduce greenhouse gases (GHG) and pollutant emissions in different road types based on improved operational parameters. Therefore, CAVs were assumed to behave like eco-driving agents to influence the environmental performance of overall traffic. A microscopic traffic and emission model platform was applied to simulate a European Medium-sized city during the morning peak period. Four roadways sections (one motorway, one rural and two urban) were selected to evaluate in detail the impact of CAVs in different roads types and over different CAVs penetration rates. Depending on the type of road, the impact may be beneficial in terms of carbon dioxide (CO2 ) and nitrogen oxides (NOX) reduction (up to 20%), but may also be negative in more congested roads.

ABSTRACT. While recent studies demonstrate the societal and economic benefits of driverless vehicles, little is known about the emission impacts of autonomous vehicles (AVs) in the context of mixed traffic. This paper explores the environmental impacts of AVs along an urban freeway corridor in a metropolitan area using Vehicle Specific Power (VSP) and EMEP/EEA emission methodologies paired with VISSIM traffic model. Three different AV penetration rates were implemented for through traffic along a freeway corridor in the city of Porto (Portugal) by considering long-term market predictions (10%, 20% and 30%). Afterwards, these scenarios were compared to current situation in terms of carbon dioxide, carbon monoxide, nitrogen oxides and hydrocarbon emissions, and travel time and stop-and-go situations. The emissions and traffic performance of each scenario were evaluated on three levels: a) overall study domain; b) corridor; c) impact of AVs on conventional vehicles (CV). AVs yielded small savings in emissions in the overall study domain for automation levels below 30% (differences in traffic performance and emissions were not statistically significant). Corridor-level analysis showed decreases of 5% in emissions can be expected with AVs technology, but penalizes travel time up to 13% for both AV and CV, when compared to the existing situation.