ABSTRACT. The next decades will be characterized by greater investments on sensor technologies and Intelligent Transportation Systems, to facilitate the paradigm shift towards full automation and connectivity in transport and mobility. If intelligent vehicles and the smart mobility services will partly mitigate the random nature of human factors, not everything in the future will be predictable. Demand flows will still unavoidably be driven by personal mobility needs, and travel choices will, on the contrary, become more complex and ill predictable due to an increasing number of multimodal and interacting sharing options. This talk will provide an overview of the current and future challenges in capturing and modeling daily mobility patterns from various sources of (big) data (GSM, floating car data, smartphones, etc.), and proposes a list of ingredients in both models and technologies, which are deemed necessary to estimate dynamic demand flows that are consistent with the observed daily activity-travel behavior.

ABSTRACT. Modelling the within-day dynamic of public transport systems is crucial for optimal real-time operations and predictive passenger information, as well as for off-line planning of the service.
In these contexts, the classical frequency based models do not match some crucial requirements, such as: complying with strict capacity constraints, representing the congestion of passengers and vehicles at stops, providing the loads of single runs. 
Schedule based models offer definitely o more proper option for operation, but are not suited to reproduce congestion phenomena that affect travel times, because they are founded on the assumption that timetables are reliable.
Some efforts have been made in the past to cope with single issues, but a comprehensive framework where to represent in a macroscopic model all relevant phenomena is still missing. The only available methods are based on agent microsimulation, which presents some intrinsic limits of stochasticity, complexity and calibration.
In this paper we present a new dynamic assignment model for transit and pedestrian networks that is capable of representing single runs, but avoids to introduce a diachronic graph, relying on a spatial/functional graph.
More specifically, we here extend to public transport the framework of Link Transmission Models, so far applied only to road networks.
The core of the proposed TLTM is the node model, which aims to reproduce congestion occurring at bus stops and rail platforms, such as: passengers that fail-to-board or fail-to-sit due to vehicle overcrowding, vehicles queuing to serve a stop, doors opened longer to allow passengers alighting and boarding.

ABSTRACT. Demand-responsive transport systems diversify local public transport with providing more flexibility of routing and scheduling and maximizing vehicles’ occupancy. The classical Dial-a-ride problem is often resolved as the minimization of a cost function. However, the projected itinerary of the vehicles is calculated on a global approach considering automatically each new users independently and a dispatcher test a match with all available service vehicles. This research proposes the conception of a mobility system based on transit vehicles and station infrastructures with exogenous constraints on the route. The vehicles are shared dynamically and make extra-stops to collect travelers with various destinations. In addition, the users of the service have not any transfer and are transported to their destination stations. We constructed an agent-based model which maximizes the utility of each vehicle while satisfying the main constraints: minimum travel time and waiting time of the users and maximum occupancy of vehicles. We propose then a new formulation of the problem. Passengers’ assignment to vehicles is controlled directly by the vehicle drivers on an aggregated level by origin destination pair. A case study illustrates the theoretical model and examines the effectiveness of the hypothesis through a sensitive analyze on operating choice as vehicle speed or on the demand ratio.

ABSTRACT. Mathematical and simulation models of systems lay at the core of decision support systems, and their role become more critical when more complex is the system object of decision. The decision process usually encompasses the optimization of some utility function that evaluates the performance indicators that measure the impacts of the decisions. An increasing difficulty directly related to the complexity of the system arises when the associated function to be optimized is a not analytical, non-differentiable, nonlinear function which can only be evaluated by simulation. Simulation-optimization techniques are especially suited in these cases and its use is becoming increasingly used in traffic models, an archetypic case of complex, dynamic systems exhibiting highly stochastic characteristics. In this approach simulation is used to evaluate the objective function, and a non-differentiable optimization technique to solve the optimization problem. Stochastic Perturbation Stochastic Approximation (SPSA) is one of the most used of these techniques.
This paper analyzes, discusses and present computational results of two cases, the application of this technique to the bi-level optimization approach to estimate and adjust time-sliced origin-destination matrices, and the application to the calibration of traffic simulation models. Variants of the SPSA replacing the usual gradient approach by a combination of projected gradient and trust region methods have been used in these studies. A special approach has been analyzed in parameter calibration cases when each variable has a different magnitude.

ABSTRACT. The transportation of dangerous goods represents one of the most critical risk particularly in urban area. For example, every day large quantities of hazardous substances have transported by roads or railways inside areas with high density of population. Thus, to ensure urban and public safety the first challenge that we tackle is to find an environmental information system which is able to provide risk management and monitoring. Moreover, it can capture urban transportation dynamics and offer specific services, visualization, analyses and evaluation of the hazmat risk in various high risk zones. In addition, due to wireless technologies and real time intelligent sensors, the risk of dangerous goods can be carried out in real time. 
In this work, we adopt a microservices based architecture to have a real-time environmental sensors system that have highly scalable applications on cloud environment. By definition, this architecture consists of a set of loosely coupled and independently deployable services. In addition, the proposed system is conducted with a smart data collector, visualization abilities and a variety of distributed sensors in order to enhance the hazmat transportation network monitoring
Finally, we show how to integrate efficiently into the system prototype the hazmat routing container embedded microservice, and illustrate the orchestration with others microservices to improve this system with hazmat routing capabilities using various routing algorithms particularly bi-directional A* algorithm. This work constitutes the first steps to build the Hazmat planning aid system that provides valuable data and knowledge to urban planner and decision makers.

ABSTRACT. Implementation of Congestion Pricing, which is one the mechanisms of transportation demand management, has so far been successfully used throughout the world. In this study, it was aimed to measure the applicability of Congestion Pricing concept in Istanbul, Turkey. Taksim district of Istanbul was proposed as the implementation area. This is because high levels of congestion are being observed in Taksim. To understand Responses of residents living in Yesilkoy district of Istanbul were collected using stated preference surveys. Yesilkoy is chosen because it is close to the Istanbul Ataturk International Airport and hence is considered as a transfer point between the airport and Taksim. Many trips between the airport and Taksim involve the usage of the express jitneys between Yesilkoy and Taksim. Moreover, Yesilkoy is mainly a residential area and generate frequent trips of the aforementioned jitneys. Travel patterns and behavior of people living in Yesilkoy and taking trips to Taksim and attitudes towards Congestion Pricing are investigated.

ABSTRACT. Travel passes offer unlimited access to public transport services within a given geographic area and time period, after the payment of a subscription-type entry fee. These tariff products are widely used in urban public transport networks for various reasons, generally supported by public opinion. There is no consesus, nevertheless, on whether the availability of travel passes is favourable for society as a whole, from an economic efficiency point of view. Microeconomic theory suggests that the welfare maximising price of a service equals to the marginal external social cost of using that service. In line with this principle, the literature normally associates subscription style nonlinear pricing systems with a profit maximising objective, which is certainly not acceptable for a monopolist public transport operator. It is not obvious, therefore, that travel passes cannot be replaced with more efficient usage-dependent pricing tools.

We present a theoretical model of public transport supply with crowding and endogenous capacity. The inconvenience of crowding is a major external cost of public transport usage that should be internalised to avoid overconsumption. Travelling has, on the other hand, a positive externality as well which appears in the form of increased service quality (Mohring effect) when capacity setting is endogenous. Also, the user benefits of travel passes may induce modal shift and thus alleviate the deadweight loss of unpriced road congestions. We investigate these contervailling effects in order to derive conclusions about the overall economic impact of travel passes.

ABSTRACT. The geography of fuel prices has many various implications, from its significant impact on accessibility to being an indicator of territorial equity and transportation policy. In this paper, we study the spatio-temporal patterns of fuel price in the US at a very high resolution using a newly constructed dataset collecting daily oil prices for two months, on a significant proportion of US gas facilities. These data have been collected using a specifically-designed large scale data crawling technology that we describe. We study the influence of socio-economic variables, by using complementary methods: Geographically Weighted Regression to take into account spatial non-stationarity, and linear econometric modeling to condition at the state and test county level characteristics. The former yields an optimal spatial range roughly corresponding to stationarity scale, and significant influence of variables such as median income or wage per job, with a non-simple spatial behavior that confirms the importance of geographical particularities. On the other hand, multi-level modeling reveals a strong state fixed effect, while county specific characteristics still have significant impact. Through the combination of such methods, we unveil the superposition of a governance process with a local socio- economical spatial process. We discuss one important application that is the elaboration of locally parametrized car-regulation policies.

ABSTRACT. The 24-km Dakar Toll Road was inaugurated in 2013 on schedule and below budget. It has reduced travel times by two thirds between downtown and the outskirts and is spurring the development of the conurbation. Long recognized to be an urgent priority, it took however about 30 years to eventuate, mainly because decision makers were deterred by the magnitude of the resettlement plan needed to insert this infrastructure into densely populated informal settlements. The paper explains (i) how the relocation of 48,000 people has been mitigated through a comprehensive compensation scheme combined with the construction of a new town to accommodate some of those relocated, and the upgrading of low-income settlements along the Road and (ii) how this US$800 million project, first greenfield toll road in Sub-Saharan Africa (excluding South Africa), and structured as a PPP, materialized through funding from multilaterals, government, commercial banks and the private sector.

The paper describes how compensation of affected households, commercial buildings, etc. have been estimated, making sure all those affected end up in at least an equivalent position as they were in prior to the project. It also explains how a new city was designed to accommodate 20,000 persons, with schools, health centers, paved roads, community facilities, running water, electricity, market, sewage, etc. In addition, it details the rationale for providing consolidated settlements along this road with better urban roads, community facilities and drainage structures, to ensure that those who are too poor to drive, would still benefit from the project.

ABSTRACT. In urban areas parking search is a considerable problem, which affects car travel time, especially in city centres where on-street parking lots are usually few compared with demand. The literature in this field mostly investigates this problem from a strategic point of view studying the impacts assessment of parking policies neglecting the problem of reproduction user behaviour in park search and the relative time spent to find a parking place at destination, which can deeply affect car travel time modelling and estimation. Nowadays, this failure plays a key role fro user information, both real-time and off-line, because the underestimation of car travel times leads to overestimate car mode attractiveness with respect to others. Moreover, ignoring parking search time implies considerable approximations in assessing transfer times and waiting for public transport in multimodal journeys, especially in case of low-frequency services. 
This paper focuses on modelling on-street parking search time by using FCD data coming from probe vehicles. It is based on data detected by probe vehicles, and identifying the typical spiral around the destination that vehicles perform in the final part of the trip to find a parking place. The proposed methodology is suitable to be used both in real-time to support user information and off-line to assess transport plans.
A real-size application to the city of Rome is presented to show the promising results obtained for a better transport modelling in urban areas.

ABSTRACT. A set-theoretical approach is presented for a multi-objective control design of the local ramp metering problem. Two control objectives are specified: first, the optimization of traffic performance, by the minimization of total time spent. Second, the emission factor of CO2 is minimized. The optimal state for traffic emission however lies in the unstable domain of the dynamic system. To dissolve this inconsistency, the control problem is formalized by using set-theoretical methods. For this purpose, the non-linear model METANET is rewritten in a shifted coordinate frame with a parameter-varying, polytopic representation. Bounds on state-, input- and disturbance variables are expressed by convex polytopes. These sets are then used for the design of an interpolated H∞ controller that is capable of improving traffic conditions according to the prescribed multi-objective criteria. Different control allocation methods are compared with non-linear model predictive control stategy, in order to illustrate the proposed methodology.

ABSTRACT. In the near future, autonomous vehicles will face with new challenges in several fields. One of the most exciting changes will be represented by the network-wide optimal traffic control. When driverless vehicles take over the road, classical road signalization schemes will become superfluous. Accordingly, the paper’s aim is to propose a control design methodology for autonomous vehicles in urban traffic network by considering the network-wide performance. The proposed two-level control strategy solves a tractable optimization problem for a network wide traffic control. On the one hand, a local intersection controller is designed which ensures safe crossings of vehicles and aims to reduce traffic emission in the junction area. On the other hand, the local controllers also optimize the network performance by minimizing the queues in all road links. The traffic is therefore modeled in a two-level fashion. A microscopic dynamics is considered in junctions and macroscopic model is applied for the whole traffic network. The control strategy is tested and evaluated based on microscopic traffic simulation.

ABSTRACT. A model-based traffic state estimation approach is developed for per-lane density estimation as well as on-ramp and off-ramp flows estimation for highways in presence of connected vehicles, namely, vehicles that are capable of reporting information to an infrastructure-based system. Three are the basic ingredients of the developed estimation scheme: (1) a data-driven version of the conservation-of-vehicles equation (in its time- and space-discretized form); (2) the utilization of position and speed information from connected vehicles' reports, as well as total flow measurements obtained from a minimum number (sufficient for the observability of the model) of fixed detectors, such as, for example, at the main entry and exit of a given highway stretch; and (3) the employment of a standard Kalman filter. The performance of the estimation scheme is evaluated for various penetration rates of connected vehicles utilizing real microscopic traffic data collected within the Next Generation SIMulation (NGSIM) program. It is shown that the estimation performance is satisfactory, in terms of a suitable metric, even for low penetration rates of connected vehicles. The sensitivity of the estimation performance to variations of the model parameters (two in total) is also quantified, and it is shown that, overall, the estimation scheme is little sensitive to the model parameters.

ABSTRACT. Walkable neighborhoods allow a variety of activities and opportunities for social encounters, through improved accessibility to commerce, public services and other facilities. Non-motorized mode choices (walking and cycling) have been promoted by governments and cities due to the substantial body of evidence on their contributions to environmental, social and economic aspects of urban sustainability. However, non-motorized transport has not been promoted in the planning and design of many cities in Israel. In light of this, and in the context of the crisis condition of the Israeli housing market, it is of interest to  examine the housing preferences for walking distance to land uses in Israeli neighborhoods. A stated-preference analysis was conducted using a choice-based conjoint analysis, based on data collected from 184 respondents of a cross-sectional, online survey. The results indicate an overall preference toward apartments that are in a walkable distance from a commercial street, rather than a shopping mall. Similarly, apartments in a walkable distance from education facilities were preferred over apartments in a walkable distance from medical clinics. Ultimately, apartment location and its price affect choice in a similar magnitude. Based on the findings, the paper provides some policy recommendations, and discusses generalization of these findings to other countries. 

ABSTRACT. Taking into consideration the role of the transportation sector in terms of energy consumption and environmental impacts, the characterization of vehicle use and driver behavior opens new opportunities for energy and emissions savings. The use of information and communication technologies can potentially be a powerful driver to promote change in this sector. 
Considering this, the objective of this research work was to identify driving behavior patterns for several driving contexts (based on hierarchical street level and weather conditions) from real-world driving data and to assess their impacts on energy consumption. The case study for this work was the city of Lisbon, where driving data from 46 drivers were collected with on-board data loggers for at least 6 months. 
The analysis performed in this work provides an insight on the impacts of driving context on driving behavior and consequently on energy consumption. Both infrastructure characteristics and weather conditions were found to cause a speed reduction and an energy consumption increase. Rain intensity was found to increase energy consumption up to 16%, while regarding infrastructure characteristics, for level 4 streets, energy consumption is 54% higher than for level 1 streets. Results provide evidence that drivers tend to drive more calmly (lower speeds and acceleration patterns) for higher rain intensities compared with dry weather. However, more local streets (level 2, 3 and 4 streets) are the ones that present more aggressive driving patterns (in terms of acceleration).

ABSTRACT. When signal changes from green to yellow at a signalized intersection, often drivers are caught in two minds whether to cross the intersection or stop at the stop line. Which might lead to right angle collision or rear end collision, when a driver makes an erroneous decision. The area in which the driver is caught in two minds is termed as dilemma zone or indecision zone. Understanding the driver behaviour within the indecisive zone can help in improving the efficiency and safety at the intersections. Most of the studies in the literature have investigated the dilemma zone driving behaviour in homogeneous traffic conditions. Traffic in developing countries like India is heterogeneous containing various vehicle types, where vehicles vary in their physical dimensions and dynamic characteristics. Heterogeneity results in the variability of approach speeds among vehicle types and the acceleration/deceleration capability of the vehicles. Driver behaviour is much more complex in mixed traffic conditions and the research findings cannot be transferred/applied directly. Study on the existence and distribution of the indecision zone at heterogeneous traffic conditions can give great insights into developing various engineering countermeasures. This study analyzes the stop/go behaviour of the drivers at the signalized intersection under mixed traffic conditions from a video based field study. The study findings will be useful in enhancing the safety and efficiency of the signalized intersections.

ABSTRACT. Driving style concerns the way a driver chooses to drive, and depends on physical and emotional conditions of the driver while driving. In order to validate this hypothesis, we propose a SEM aimed to investigate on the relationship between among driving style and drivers’ characteristics such as somatic, behavioural and emotional conditions. Drivers’ conditions include tiredness, sleepiness, sickness, gloom, worry, nervousness, boredom, and anger. In the proposed model, driving style is considered as an endogenous latent construct, while drivers’ characteristics were considered as exogenous. Driving style is defined by means of a judgement expressed by the driver on a scale ranged from aggressive to cautious. In addition, a more reliable definition of the driving style is determined through an objective measure derived from cinematic parameters. We addressed to a sample of drivers a questionnaire aimed to collect information about their conditions while driving and the judgement about their driving style. Each driver registered the same path run in different days (more than thirty paths for each driver, in average), and complete the questionnaire for each path. This permits to observe the possible changes of drivers’ driving style as a function of the different physical and emotional states that drivers present in different days.

ABSTRACT. Decision makers need appropriate information systems to carry out spatial analyses related to transportation. Hence, Geographical Information Systems (GIS) are one of the effective tools used for realizing spatial analyses. Therefore, GIS were also used in this study for evaluating the information about intercity bus travels in Turkey. For this purpose, the intercity bus travel matrix obtained is transferred to GIS environment. Thus, it was possible to evaluate the intercity bus travel matrix in a visual environment.
Several methods are used to obtain Origin-Destination (O-D) matrices. However, it is generally expensive and time-consuming to use classical methods based on home surveys or roadside interviews. Therefore, other methods that are not expensive and time-consuming have been developed. The model developed by Bell is also one of them, and estimates travel matrices from traffic counts. In order to find the daily intercity bus travel matrix for the year of 2013 in Turkey, this model was used. 
All of the 81 provinces in Turkey were considered in this study. The highway network constructed consists of 714 links, 268 nodes and 81 regions. All or nothing assignment method was employed, and the distances among cities were taken as the cost parameters. Besides, the principles of the Gravity Model were used to form the initial matrix related to intercity bus travels. The calibration constants of the Gravity Model were obtained from the multiple regression analysis based on the daily bus runs that were made from Istanbul and Ankara to some cities reciprocally.

ABSTRACT. We propose an optimization framework for urban bicycle network design. The model takes into account interests of the users (who travel along shortest paths) and the planners (available budget). An underlying network composed by street segments suitable to build cycling infrastructure is taken as input. Each network link has construction and user cost, both proportional to the distance. A network link without cycling infrastructure which is part of a path followed by users, has a larger user cost. A multi-commodity network flow mixed-integer mathematical program is proposed and applied to small-sized problem instances to validate the model. The formulation considers the discontinuities of the bicycle network, i.e. the users' paths which include segments without cycling infrastructure. Sensitivity analysis are performed with respect to budget levels and to penalization of user's cost in links without cycling infrastructure. A metaheuristic is proposed to handle large-sized instances. As an additional feature (difficult to formulate in the exact model), the metaheuristic also minimizes the total number of discontinuities by including them into the objective function. The accuracy of the metaheuristic is estimated by comparing with exact results when possible. The methodology is tested using data from the city of Montevideo, Uruguay, including a large-sized underlying street network and origin-destination trips estimated from a household survey. Computational results are obtained with and without minimization of discontinuities, and they are compared with the current bicycle network of the city.

ABSTRACT. The planning of pedestrian crossings in urban environments could benefit from the use of microsimulation tools, which can reproduce the movement of road users and estimate the Level of Service of all traffic modes. With regard to pedestrian motion, the LOS is usually assumed as a function of the delay time to cross the road – HCM has identified threshold values of delay time to assess LOS. However, this approach may be inadequate in areas where pedestrian motion has to be promoted, like in the proximity of pedestrian zones and squares where recreational activities are possible. 
In this work a new holistic indicator for pedestrian LOS is used to evaluate quality of pedestrian motion for different types of crossing facilities. The indicator considers three aspects of pedestrian motion, namely efficiency, safety and comfort, and has been developed in the past research [Pascucci and Friedrich, 2016]. The final aim of the current work is actually to test it in the overall evaluation of a pedestrian crossing facility, by examining the relation between motorized vehicles and pedestrian LOS in different design scenarios and traffic conditions.
The estimation is firstly carried out on a real case study, namely the shared space in front of the train station of Bergedorf in Hamburg. Successively, other alternatives are implemented and simulated in VISSIM, e.g. traffic light, pedestrian refuge and zebra crossing. Best design alternatives are finally discussed by the analysis of the resulting LOS of pedestrian and motorized vehicles.

ABSTRACT. The highway capacity manual of US (HCM-2010) is widely referred document around the globe for planning and design of roads. However, the transferability of HCM recommendations for heterogeneous traffic conditions represents a major research issue as they are often not suitable to properly explain the traffic complexities of a mixed-traffic condition. This research investigates the suitability of HCM equations for determining the entry capacity of a four-legged roundabout under mixed traffic condition and proposes a methodology for validating and calibrating HCM equations for performance evaluation. Data corresponding to traffic composition, traffic volume, critical gap and follow-up time have been extracted from the video records collected at two roundabouts. The relationship between entry flow and circulatory flow has been plotted from the observed data when there is a complete saturation in the approach leg and depicted that it follows a negative exponential behavior. It implies that the entry capacity reduces exponentially with the increase in circulating flow. The critical gap has been estimated by various methods such as Maximum Likelihood Method (MLM), Root Mean Square (RMS) method and Probability Equilibrium Method (PEM). Further, a stream equivalent critical gap of 1.60 seconds and stream equivalent follow-up time of 1.24 seconds have been derived due to the mixed traffic scenario at the study location. A multiplicative adjustment factor of 1.1 is suggested for the use of HCM 2010 equation directly to estimate entry capacity under heterogeneous traffic condition. The study results may use for planning and designing of roundabout under mixed-traffic flow condition.

ABSTRACT. Nowadays plenty of navigation and route guidance methodologies are available, based on real-time traffic information collected from technologies that have not been originally developed to measure road traffic parameters, for example, cellular and GPS data of users’ mobile phones and their current travel demands. Traditional traffic estimation methodologies, such as the four-step model, are ignored. This way, reliable traffic data can be obtained only from those areas where there are enough users.
In our paper, we present a route guidance methodology that combines current transportation demands with the results of the traditional four-step model. The predicted traffic state of the network is calculated for every fifteen minutes of the day by using a dynamic assignment with predefined static demand matrices as a first assignment. When travelers use the route suggestion system, their demands are collected in an actual demand matrix for the same time interval. This matrix is then combined with the original static demand matrix for this period and assigned to the network as a second assignment. The real-time traffic disruptions on the network are also taken into account. Users will be provided with route suggestions based on the combination of the results of the two assignments.
The methodology is tested based on real static demand and network data of Budapest, using an integrated multimodal transport model maintained by the BKK Centre for Budapest Transport. The actual demands are simulated, modelling various traffic situations.

ABSTRACT. Stochastic User Equilibrium (SUE) has been proposed to overcome the limitations of the perfect knowledge assumption of Wardrop’s user equilibrium. Imperfect knowledge of the network and the associated heterogeneity in the perception of travel times, as it occurs in the absence of Advanced Traveller Information Systems (ATIS), justifies SUE providing strictly positive flows for all routes. SUE is at the same time the steady state of a day-to-day assignment process under the assumption of state independent route choice, i.e. choice independent of the previously chosen route. Still in a context of absence of ATIS, state-dependent route choice represents a more realistic assumption: users exhibit inertia to change. Experimental evidence supports the assumption. The paper investigates the fixed-point states of Markov assignment processes with state-dependent route choice. These fixed points characterise a new concept of network equilibrium, referred to as State-Dependent Stochastic User Equilibrium (SDSUE), where if each user shifts from her current route to her newly chosen route the observed route flows do not change. The existence of SDSUE is guaranteed under usually satisfied conditions. A method of successive averages is proposed for computing SDSUE. An example related to the Braess network with logit route choice illustrates the model. Numerical evidence is provided of a symmetry property of SDSUE transition flows whereby, at equilibrium, the number of users shifting from route i to route j equals the number of those shifting from route j to route i.

ABSTRACT. Stochastic optimization algorithms have been proposed in the recent literature as a preferred way for calibrating Dynamic Traffic Assignment (DTA) models, because the computation of explicit gradient is numerically too cumbersome on real networks. However, early experiences based on SPSA have shown convergence issues when the number of variables becomes large.
This induces to focus on some structural demand variables rather than to consider all components of origin-destination (O-D) matrices. Moreover, with the possibility of distributed computing, many algorithms that where not efficient in a standard configuration (i.e. sequential objective function evaluation within each iteration) can become a viable alternative to SPSA. For example, parallelization is especially beneficial for genetic algorithms, which require a large number of independent function evaluations per iteration.
In this paper we examine several optimization algorithms applied to dynamic demand calibration based on flow and speed field measurements. The problem is to minimize the distance between results of a dynamic network loading and traffic data observed on road links. Flow data, obtained through various sensors at given points, are standard measurements on the road network today. However, speed data, obtained from GPS devices, are becoming more accessible and can provide area-wide measurements.
Several test networks and some real-world scenarios are investigated in the context of laboratory experiments, where known O-D matrices are perturbed after its dynamic assignment on the network, to prove the scalability of the proposed methodology as well as its effectiveness. The preliminary results indicate which algorithms can be successfully applied in this context.

ABSTRACT. Network capacity in a transportation system becomes an important measurement for transport planning and management because it addresses its capability to satisfy an efficient network traffic flow reducing the inefficiency of congestion phenomena.
This work provides a discussion of road urban transport network capacity including existing definitions in literature and the validation of new measurement methods. The study explores some of the properties of network-wide traffic flow relationships in a large-scale complex urban street network using real-time simulated results obtained from a dynamic traffic assignment model periodically updated by data from radar sensors through rolling horizon technics.
The basic variables used in the methodology, such as network flows, speeds and lengths are characterized using a network model calibrated in the urban area of Catania (Italy). For a comprehensive yet simple analysis, equations and graphs are utilized to resume the obtained results related to different days and several time intervals of the day.
This procedure proved to be suitable to investigate the properties of network-level traffic flow relationships and concluding remarks include suggestions for further research in this highly promising area.

ABSTRACT. This paper proposes a novel approach to investigate the community structure in a city mobility network using multi-layer graphs of network traffic that represent the flow of different movement entities (i.e. private vehicles, buses and passengers) in the network. Given a city network and link weights that define the strength of connection between different parts of the city, we partition the area into smaller regions at each layer. The partitioning algorithm returns surprisingly compact and connected regions in all layers. In addition, we define a similarity measure that compares community structures across the layers and quantifies the level of interaction between them.

ABSTRACT. Nowadays, data sets are spreading continually, generated by different devices and systems. The modern GPS based tracking systems and the electronic tickets are producing lots of data, and we could use them, for improving the service level. These data are processable with the modern devices and methods, and we can use them for obtaining information. Thanks to the spread of data mining, these tools are not appearing only in marketing research, but also in the most various kind of scientific areas and they are advertising a new scientific revolution. Although the importance of these data sources is essential it is not widespread in transport planning except in some specific areas.
The smart card systems store the number of boarding passengers and in some cases also the alighting values. From the passengers’ boarding and alighting information in a stop point we can create a time series, which shows the behavior type of the given stop points presented on graphic curves. With the help of different clustering and classification processes, these curves can be turned into groups and we can observe these groups of stop points which are defining separated zones. This is the basic step in transport modelling and the zones were determined by manual methods usually.
In this paper we examine clustering and classification methods compared to each other and check the usability of different distance measurement techniques. This paper shows the usage of these methods in public transportation and presents the background of this kind of zone distribution technic.

ABSTRACT. Disruptions in public transport have major impact on passengers and disproportional effects on passenger satisfaction. The availability of smart card data gives opportunities to better quantify disruption impacts on passengers’ experienced journey travel time and comfort. For this, accurate journey inference from raw transaction data is required. Several rule-based algorithms exist to infer whether a passenger alighting and subsequent boarding is categorized as transfer or final destination where an activity is performed. Although this logic can infer transfers during undisrupted public transport operations, these algorithms have limitations during disruptions: disruptions and subsequent operational rescheduling measures can force passengers to travel via routes which would be non-optimal or illogical during undisrupted operations. Therefore, applying existing algorithms can lead to biased journey inference and biased disruption impact quantification. We develop and apply a new transfer inference algorithm which infers journeys from raw smart card transactions in an accurate way during both disrupted and undisrupted operations. In this algorithm we incorporate the effects of denied boarding, transferring to a vehicle of the same line (due to operator rescheduling measures as short-turning), and the use of public transport services of another operator on another network level as intermediate journey stage during disruptions. This results in an algorithm with an improved transfer inference performance compared to existing algorithms.

ABSTRACT. New data collection technologies have all, but replaced traditional, site-based data collection methods for trip matrix development in the UK. Event data produced by Mobile Network Operators provides adequate information to identify time and location of trip origin and destination (OD), but only limited information on transport mode, vehicle type or trip purpose. In addition, GPS-based data provides another source of OD and user data.

This paper describes the mobile phone and app-based demand data collection, verification and processing toward the multi-modal strategic transportation model of Chelmsford, UK. The mobile data was collected during 30 days by INRIX, and was processed by Jacobs to develop time-of-day, vehicle type, mode and purpose-specific motorised trip matrices. The mobile phone- trip matrices were verified and adjusted against various third party data.

A cycle app was developed for Chelmsford to facilitate non-motorised OD data collection by engaging a voluntary user group. Together with other data sources, the data was used to develop cycle OD matrices and to calibrate a cycling route choice model.

The paper concludes that the use of mobile phone data provides greater quantity and coverage to develop OD matrices. However, the new data source requires significant effort in post processing and model data segmentation by mode, trip purpose and vehicle type, via applying third party information. The quality and efficiency of this process is likely to improve in the near future, resulting in a step change in the modelling practice. The model was delivered to Essex County Council in April 2017.

ABSTRACT. Due to varying traffic volumes and limited traffic infrastructure in urban areas, travel times are uncertain and differ during the day. In this environment, city logistics service providers (CLSP) have to fulfill deliveries in a cost-efficient and reliable manner. To ensure cost-efficient routing while satisfying promised delivery dates, information on the expected travel times between customers needs to be considered appropriately.

Typically, vehicle routing is based on information from shortest paths between customers, to determine the cost-minimal sequence of customer visits. This information is usually precomputed using shortest path algorithms. Most approaches merely consider a single (shortest) path, based on a single cost value (e.g., distance or average travel time). To incorporate information on travel time variation, it might be of value to consider alternative paths and more sophisticated travel time models such as Interval Travel Times (ITT). 

In this work, we investigate the incorporation of alternative paths into city logistics vehicle routing. For this purpose, we compare our approach to classical shortest path approaches within a vehicle routing problem. Our approach considers a set of alternative paths and incorporates ITT. Experiments are conducted within an exemplary city logistics setting. Computational results show that the consideration of alternative paths allows to select better paths with regard to a trade-off between efficiency and reliability when travel times are varying.

ABSTRACT. Journey planning is a key process in public transport, where travelers get informed how to make the best use of a given public transport system for their individual travel needs. For a given journey request, the now ubiquitous journey planner applications usually offer one or several routes, which are linear sequences of legs forming the itinerary. A common trait is that they assume a deterministic environment. However, vehicles in public transport often deviate from their schedule.
In addition, there are often multiple alternative services which a passenger can choose from at a given location. Despite providing some adaptability, linear journey plans are not able to capture the full amount of flexibility inherent in multi-service public transport systems.

To fully exploit the given flexibility and to pre-plan adaptive decisions accordingly, we propose the concept of a routing policy instead of a linear journey plan. It is a state dependent routing advice at each location, consisting multiple services and specifies exactly which service to take in which situation. The traveler may define an arrival time dependent utility value at the destination. The goal is to find a policy maximizing the expected value of the utility achievable by following the policy.

This work deals with both arrival time and current time dependent policies. We present polynomial algorithms for the cases when the number of recommended services at each stop is bounded by a given number k. Then, an extensive comparison is presented showing the achievable gains when applying the different policy types.

ABSTRACT. Bluetooth technologies can be used to track specific vehicles by their MAC address; this can be done by installing Bluetooth detectors in selected intersections of a study area. Based on the detections, it could be possible to track vehicles equipped with this technology. However, an equipped vehicle is not necessarily detected at every intersection as detection probabilities depend on many factors such as weather conditions, nearby infrastructure or vehicles speeds. Given the lack of perfect information, it is necessary to infer the most likely routes chosen by the vehicles.

This study presents a methodology to infer route choices in a network by reconstructing paths between two successive detections of the same vehicle. An estimation of the probabilities is obtained for each route between these detections points. Once these probabilities are obtained, it is possible to infer the entire route from vehicle’s origin to its destination. Being able to predict these travel patterns is essential for transport planning and different polices such as traffic management and route pricing.

The methodology has two steps. In the first step, with the collected Bluetooth data, a distribution of the time spent by the vehicles at each intersection is calibrated. Additionally, travel time distributions between intersections are obtained. Both types of distributions are calibrated for different periods of the day. The second step consist in convoluting the previous distributions between two successive detections to obtain aggregated distributions for each potential route. Based on the observed travel time, Bayesian inference is performed over the aggregated distributions.

ABSTRACT. Major cities encounter traffic problems every day, whereby studies showed that cars looking for a parking spot have a large impact on urban traffic. 
This paper presents an approach for a park spot route in a city using on-street parking information. The result is a route through streets with high parking probabilities close to the destination, where the driver decides where to park their car. 
The proposed method is based on the A* algorithm, a shortest path algorithm developed from the Dijkstra algorithm. The A* algorithm inherits the easy implementation and possibility for adaption from Dijkstra, but has a shorter computation time. 
For the park spot route the cost function of the A* is adapted in that sense that it not only takes the travel time on a road segment, but also the parking probability on this segment into account. The main part of the paper is constituted by the discussion on how to find a suitable cost function that limits these two variables into one common interval so that they have the same impact on the route choice.
The development of the presented park spot routing algorithm is based on an artificial square grid with artificial travel times, lengths and parking probabilities per road segment. 
Simulation results showed that adapting the A* algorithm leads to the right path for the park spot routing problem.
Finally, a short outlook on possible scenarios will be given, since the algorithm should be applied in different cities.

ABSTRACT. Cities have been often organized in terms of planning with special attention to motor vehicles and not well prepared for pedestrians and cyclists. In order to privilege active modes, there is the need to ensure the safety of these vulnerable road users.
The main objective of this paper is to implement a statistical analysis to assess the occurrence and severity of road crashes involving vulnerable road users. This research is focused on analyzing the trends and causes of road crashes involving cyclists and pedestrians and what are the main difficulties that people using active modes do face in their journeys. In order to reach this objective, a database of crashes registrations involving motor vehicles and vulnerable road users from Aveiro, Portugal, between 2012 and 2015 was created. This analysis intends to evaluate the evolution of the number of crashes and to create patterns of risk factors such as weather conditions, specific locations and singularities that might represent additional risk, profile of pedestrian or cyclist involved. Regarding the analyzed variables that characterize crashes participations, the dependent variables considered were: meteorological conditions, location, proximity to a pedestrians’ crosswalk and gender of the VRU.
The probability of the vulnerable road user being a pedestrian increases by 2.7 times if the crash occurs on a urban street segment, 10.6 times if the crash occurs at a pedestrians’ crosswalk, and 3.5 times if the VRU is a female.

ABSTRACT. The research aims to calibrate Injury Crash Rate Prediction Model ICRPM for two-lane rural roads focusing on single-vehicle run-off-road crashes happened on the study network during an 8-year period. Five years of crash information were considered to calibrate ICRPM while the remaining 3 years were used to validate the results. A wide variety of factors appear to influence or be associated with the crash dynamic investigated: gender, age and number of drivers, horizontal curvature indicator, mean lane width, daylight/nighttime conditions, road surface conditions, Annual Average Daily Traffic, mean speed. Before moving to the calibration phase, a mixed logit model aiming to estimate the injury severity ranging from no injury to fatality considering also other intermediate values, was calibrated. This model is particularly advantageous in terms of accounting for unobserved heterogeneity (unobserved factors across crash observations may be considered still in case of limited data). Most dangerous scenarios that can affect the severity of the study crash types were identified and adopted for calibrating generalized estimating equations with a negative binomial distribution that predict crash rate. Akaike information criterion and Bayesian information criterion were used for checking the reliability of the results. The validation procedure succeeded: the diagram of cumulated squared residuals confirmed the reliability of the model. The construction of final risk type density diagrams to thoroughly identify all possible combinations of existing explicative variables likely to produce hazardous circumstances on the road have been plotted for identifying unsafe and safe scenarios.

ABSTRACT. Sharrows are a type of lane marking used on shared-use streets to indicate that cyclists have the right to, and can be expected to cycle on the roadway. These markings are widely used in countries such as USA, Canada and Australia but have only recently been implemented in Norway. It is important to determine how sharrows function within the Norwegian context, including investigating how Norwegians understand sharrows and whether their behavior is influenced by the markings. To examine this, behavioral data was registered manually on-site before and after the implementation of the new marking. Additionally, surveys were distributed to both cyclists and drivers who use the street, in order to learn more about the users’ perceptions of the sharrows and their effect. A decrease in sidewalk cycling among adult cyclists was observed, but no significant changes in cyclists’ positioning in the roadway |were found. However, the results of the survey showed that the majority of the users understood the purpose of the new measure. Also, it was found that 69% of the cyclists reported a higher degree of perceived safety and 70% of the drivers felt that the sharrows has influenced the traffic safety along the street in a positive way. While the results of this study are not entirely conclusive, and specific to the location and conditions of the implementation, no detrimental effects were observed.

ABSTRACT. The identification of road sections characterized by high risk accidents is the first step for any successful road safety management process, considering the limited available resources. Although researchers started to study black spot decades ago, there are many un-clarified questions in this field. In the identification process of black spots three main methods can be used: screening methods, clustering methods and crash prediction methods. Many literatures and case studies were written describing each method pros or cons. These literatures concentrate mostly on one type of road each time, although road characteristics (i.e. speed, ADT) can highly affect the success and precision of the applied method. Therefore, the most important question to be answered is which method for which road?. This question can be answered by comparing different applied methods for different road types. However the comparison of different methods is still not adequately explored area. This article aims to compare different methods used in identifying black spot; the sliding window and the spatial autocorrelation for two types of roads differ in their average speed, where speed is one of the important road characteristics which is still not adequately explored. The result shows a preference to use the sliding window for identifying black spot in high speed roads and the lack of preference to use it in low speed roads, and vice versa for spatial autocorrelation method, following accidents distribution pattern. And a result of a weakness in applying Empirical Bayesian in high speed road is also included.

ABSTRACT. Airline re-fleeting, also known as Demand driven dispatch, is the reassignment of aircraft to flights close to departure in order to improve operating profitability. It swaps aircraft of different sizes on flight legs to change capacity in response to demand. With schedules fixed many months ahead of the departure date, revenue management operates with the assumption of fixed capacity to maximize revenue. Demand driven dispatch makes capacity flexible again, changing aircraft assignments nearer to departure. Using more detailed and reliable demand information to better match capacity supplied with quantity demanded, revenues may be increased and operating costs decreased. Previous research on demand driven dispatch have not incorporated or significantly simplified key issues such as revenue management and maintenance operations; making changes few days in advance to the day of operations may dramatically influence aircraft airworthiness if available flying hours until maintenance are not taken into account. The focus of this paper is on the integration of demand driven dispatch, revenue management and maintenance operations in an airline network environment. A stochastic mixed integer model is proposed in order to solve the fleet re-assignment problem while considering aircraft maintenance requirements and revenue management. Stochastic demand is considered in order to model demand predictions and to decide on seat inventory control, i.e., so as to decide how many seats to protect at each fare class and market. Realistic computational experiments drawn from IBERIA, the major Spanish airline, are presented.

ABSTRACT. Immediately after a natural disaster, there is the need for several assistance activities in order to help the victims in the affected areas. The challenges faced in this response operation are numerous, given the lack of centralized coordination, limited resources, lack of infrastructure and personnel. The logistics process is initiated by the assessment of the local situation after the emergency alert (relief) given by the authorities in the affected country, or by the international community, depending on the scale and severity of the disaster or crisis, in a period up to 24 hours after the disaster strikes. Therefore, to support this initial assessment, the use of real-time mapping and research technologies, such as Unmanned Aerial Vehicles (UAV), is important to increase logistic efficiency in the distribution processes. In this context, the objective of this article is to present a procedure to help in the decision-making process of designing an aid distribution network with the use of UAV technologies and geographic information systems (GIS). The procedure also helps in the location of points of distribution and in vehicle routing in the distribution problem. The procedure is applied in a real crisis situation, taking as a basis the characteristics of the disaster response operation occurred after the floods in the municipality of Duque de Caxias in Brazil in 2013. The results of the association of UAV and GIS enabled the evaluation of the amplitude of the disaster in region and identifying possible service facilities within the affect region as supply centers.

ABSTRACT. Airport ground access is one of the key determinants influencing air travellers’ airport choice. The improvement of transportation facilities serving airports could represent a strong strategic action to enlarge the catchment area and increase airports’ market position.
The continuous growth of air travel demand and the consequent induced road congestion have encouraged the development of efficient transit systems reaching the airport, thus promoting a modal shift from individual cars to greener transport alternatives. In addition, transit systems must be resilient and reliable to air travellers, since the cost of missing a flight is high. 
In case of disturbances to the transit system (for example strikes or infrastructural failures), it is fundamental that passengers are still enabled to reach the airport without losing their flight. Alternative and easily recognizable transit travel solutions and quick recovering from such disturbances are crucial to allow the transport of affected passengers within acceptable time. 
In this paper, several operational and modelling resilience aspects of transit systems accessing airport areas are discussed and the case of a large regional Italian airport is examined. Three different transit systems to get to the airport (Automated People Mover, Airport Shuttle Bus, Bus Line) are modelled. The interruption of one of those services and the consequent shift to other transit modes are studied. Effects on passengers are analysed, as well as the time required for the system to return to the original state. Finally, this work provides tools to improve airport ground accessibility by enhancing its reliability and resilience.

ABSTRACT. The limited capacity available at airports regularly leads to an imbalance between supply and demand at busier airports. It is the runway systems at such airports which often set limitations on the number of possible flight clearances. In addition, existing systems of regulation sometimes use strategic resource allocation as a condition for access to specific infrastructure, as for instance the airport slot allocation in accordance with EU Regulation 95/93 and its amendments in Europe.
These regulation systems do indeed restrict the number of demanded slots to a level which the capacity can, on average, handle, but this alone cannot prevent unnecessary waiting times before customers are served. One reason for this can be temporary fluctuations in the capacity on offer (e.g. due to weather influences) and another is deviations from the forecast demand (e.g. due to delayed departures or short-notice changes to flight routes of an aircraft). 
Today’s system applies a first-in-first-out system in allocating runway capacity to demand in the form of planned flights. In order to treat all flights equally, deficits in utilization of the available slots are accepted. 
The concept presented in this Paper allows user-driven priorities to be set and true-to-schedule user behaviour to be promoted. It furthermore enables prioritization of high-profitability flights during clearance in cases where there are capacity deficits.
We show which runway system prioritization is suitable in aviation to increase efficiency in resource utilization. We present KPIs which would be required by an airport to make this effect appropriately measurable.

ABSTRACT. Spatial structural analytics have increased importance in recent years as a helpful tool for decision making as well as public operators’ commitment in integrated transport planning and land use. Analysis of complex systems remains the domain of excellence to grasp for control complexity. The public transportation transit system assessment targets two important aspects of public transport network: the organizational diagnosis and the technical diagnosis. Theoretical foundations derived from order theory (combinatorial connectivity, spectral graphs analysis) provide interesting solutions to study exchanges and flows in a system. These are systemic tools well suited to big linked data analytics. Various applications exist in the areas of computer networks, social networks analysis, economic analysis and, knowledge discovery. This paper involves simplicial algebra and centrality based on spectral graphs for the discovery of the organization and the existing linkage between transport supply given by the routes served by transport (bus, tram, metro, train) and demand resulted in attractive spatial zones (stop stations generators). The case study concerns the analysis of urban transportation systems of the city of Valenciennes (France). However, the approach remains valid for other networks without any alterations or modifications.

ABSTRACT. When public authorities face the need to improve a transportation system, they normally have to make a difficult choice among a set of technological and operational alternatives. To help the correct evaluation of each alternative and its impacts, costs and benefits, it would be useful to have a decision support tool based on approaches such as Multi-Criteria Decision Analysis (MCDA) and/or Cost-Benefit Analysis (CBA). Among the many impacts caused by a public transportation system, typically those on the land use are not adequately considered in the decision-making processes, mainly because they are hard to monetize, they are often considered as value transfer instead of value creation, and they are too complex to be assessed by traditional transport modelling tools. To overcome these weaknesses, the objectives of this research are to identify and measure the impacts of transit systems on land use and accessibility, and to consider those impacts in decision-making processes, along with more traditional financial and transport related impacts. For this purpose, a decision support tool, combining a land use and transport model with a MCDA model, was developed and assessed in a small case study. In future work, sensitivity and risk analysis will be incorporated, to more accurately and realistically reflect uncertainties and exogenous conditions that may significantly affect the costs and the benefits of a project. Finally, this decision support tool will be fully assessed in a study of the Green Line extension project in Boston, USA.

ABSTRACT. Health care is one of the most important facility and it helps to improve the quality of life and social welfare of modern society. Recent advances in the field of health geography have greatly improved our understanding of the role played by geographic distribution of health services in population health maintenance. This study presents a two stage methodology to evaluate the existing locations of health care facilities in Tiruchirappalli city of Tamil Nadu, India and includes to identify the optimal site for new healthcare facility. Firstly, the Spatial Accessibility index to healthcare facilities are measured using three Step Floating Catchment Area (3SFCA) in a geospatial framework, considering three variables; namely, attractiveness of healthcare facilities, travel time or distance between the locations of the service centre and the residence, and population demand for healthcare facilities. This index would define the disparity region in healthcare accessibility in the city. Secondly, the Location-Allocation model would be developed to find optimal sites for new healthcare facility in the deprived area. The analysis also includes road network analysis, to determine the closest facility and shortest route to these health care facilities from the population. The study is thus, recommended as a spatial decision support system for urban policy makers regarding accessibility of healthcare facilities in the urban area.

ABSTRACT. In high frequency transit services, regularity is a key element, as it represents a qualifying measure of service-quality in terms of time reliability. Irregularity problems are unavoidable due to the stochastic context where bus services are operated. Therefore, characterizing the regularity and identifying irregularity causes and sources provides an opportunity to maintain planned headways.
Previous research examined the irregularity causes and sources by using scheduled and actual arrival (or departure) times at bus stops. However, to the best of our knowledge, no studies analysed the irregularity causes and sources by comparing arrivals and departures headways between two consecutive bus stops. This analysis is particularly useful when buses run with shorter headways and it is difficult to maintain the planned timetable.
To address this gap, this paper proposes an off-line framework for high frequency bus systems, to characterize the regularity over all bus stops and time periods and disclose systematic irregularity sources from collected Automated Vehicle Location (AVL) data by inferring information on headways only. Moreover, this framework selects preventive strategies, accordingly.
This framework is tested on the real case study of a bus route, using about 15,000 AVL data records provided by the bus operator CTM in Cagliari (Italy), whose vehicles are all equipped with AVL technologies. Easy-to-read control dashboards are used to represent results.
The experimentation shows that transit managers could adopt this framework for accurate regularity analysis and according to service revision.

ABSTRACT. In an era of fast technology improvements and easy access to real-time data from various sources, intelligent transport systems emerge. Emergency medical services (EMS) response is one of the transport systems that can take advantage of technological advances. Moreover, cities present themselves as dynamic environments where traffic flows change during the day as well as people’s location. Therefore, a static EMS response is inappropriate and unable to give a proper response at every period of a day with a reasonable ambulance fleet size.
This paper studies real-time algorithms for ambulance dispatching using real-time information from different sources, traffic conditions and emergency calls. We propose a methodology that incorporates a model to assess the next hours’ requirements for urban EMS accordingly to the live input of emergency medical calls. With the use of this model information and real-time traffic conditions, an intelligent will allow a better assessment of which ambulance is to be dispatched. Furthermore, we will test the proposed dispatching algorithm in a simulation model and compare with real life dispatching practices.
We apply our methodology to a case study, Porto city, to validate it and assess the impact of ITS in dynamic environments.

ABSTRACT. This paper provides an analysis of long-short distance passenger interconnectivity in the European context. The analysis is based on the results of the project HERMES (7th FP). The information about connecting trips surveys were carried out to stakeholders and travellers. Their outputs show weakness and strengths of four intermodal stations: Gothenburg Central Station (Sweden), Avenida de America Interchange in Madrid (Spain), Lleida-Zaragoza railway stations (Spain), and Part Dieu Intermodal Station in Lyon (France).

The stakeholders’ surveys highlighted management features and characteristics of interchanges. The survey conducted to passenger gives an insight into the key requirements of long-short distance intermodal services. Passenger surveys provided information about the trip and their socioeconomic characteristics. In addition, they rated the importance and satisfaction of a series of aspects.

The paper identify the most relevant elements of each interchange -their weakness and strengths. These findings consider both providers’ and customers’ perspectives. The most common weakness in terms of management is the lack of internal coordination among operators, managers and decision makers, which influences the quality of the information provided to passengers. The strengths of each interchange depend on the customers’ personal profile. In some cases, the availability of a variety of cheap urban transport services is the most valued characteristic. In other cases, customers prefer good quality and comfortable facilities.

ABSTRACT. Nowadays, one of the main functions of logistics is material flow distribution from the producer to the final consumer. Modern scientific methods and models pay attention to numerous tasks at different stages of the logistics system. But, simultaneously, the efficiency of interaction of logistics systems considered in isolated form from the problems of improving the whole society efficiency. Meanwhile, there is a connection between the current expenses of trade organizations, industrial producers and consumer spending. All this lead to necessary of include consumer’s component into logistics system. The development of logistics approach based on consumer’s cost-benefit opens new opportunities to improve their service and the efficiency of the logistics system. The article examined influence of consumer on the logistics system efficiency functioning; Defined interaction of logistics system and consumer; Revealed prerequisites that determine necessity of considering consumers in logistics system; Proposed models that describe integrated efficiency evaluation of the logistics system and customers.

ABSTRACT. The future holds great promise for the use of autonomous vehicles. Many daily activities which require the use of extensive manpower, will soon be performed with much less human interference, if at all. City logistics is an example of one such arena, which will be highly affected by the introduction of autonomous vehicles. Autonomous vehicles will replace human dispatchers in the near future, and conduct delivery tasks. However, this process will be gradual. Until granting full autonomy, autonomous vehicles will be integrated in the general traffic in the form of platoons, led by a human driver. 
In this paper we examine the use of platooning for delivery purposes in a micro-simulation environment. In our developed model, autonomous vehicles perform delivery tasks, traveling from an origin depot to a customer, by joining platoons along the way, which bring them closer to their destination. This model relies on the use of autonomous connected vehicles, enabling the autonomous vehicles to travel in the form of a platoon, and the communication between vehicles, enabling vehicles to change platoons along the way. 
The proposed setting raises questions with respect to various aspects concerning the platoons, as changing their size (joining or reducing vehicles) and where these changes can occur. Additionally, aspects affecting the general traffic should also be considered, as the operation of platoons in signalized intersections and their impact on the travel time of the general traffic. These issues are investigated using the developed micro-simulation, and conclusions are drawn concerning the optimal operation mode.

ABSTRACT. Traffic along a transit line involves two kinds of mobile entities: passengers versus vehicles. The paper develops a stochastic model to deal with: headways between successive runs serving stations, wait times at boarding stations, passenger flows per vehicle and by leg (i.e. pair of entry-exit stations), in-vehicle comfort differentiating between seated and standing places. The “Rank conservation” postulate of Leurent et al. (2012a,b) is used to establish analytical formulae for the expectation and variance of every traffic variable of interest: leg flows, link flows, wait times, leg physical times, generalized times at waiting and in-vehicle. The linkage between waiting prior to boarding and in-vehicle crowding is modeled: each user is concerned individually, conditionally to the headway during which he waits for the vehicle. A computation scheme is provided to deliver the statistical summaries of the array of traffic variables.

ABSTRACT. The purpose of this paper is to show relevant dimensions for presenting railway punctuality data. We discuss alternatives for visualisation and analysis and show principles and practical examples. The study has benefitted from access to a complete database of Norwegian punctuality data including all data that have been automatically stored since the system was introduced in 2005. The applied data include information about train number, scheduled departure and arrival to each of the stations, as well as actual departure and arrival times. In addition, we have had access to infrastructure data from the Norwegian railway authority. Based on the available data, we developed a set of visualisations and analysis options. The tools were first implemented as a prototype, and then transferred to the Norwegian Railway Authority to be hosted on their web pages. This paper describes the principles applied in the tools, as gives a wide range of illustrations of actual implementation of the analyses. Three key dimensions in punctuality analysis are related to the line, time and selected trains. We use several tools that combine the time and line perspectives, including heat maps and correlations analysis between punctuality on different locations or points in time. Both experience-based knowledge and literature indicate a need for a practical method designed for use by practitioners in the railway sector. The main outcome of this research is the creation of an analysis tool case to meet these needs. The visualisations support informed and efficient identification of punctuality improvement measures.

ABSTRACT. Nowadays, the railway control and planning methods incorporate the robustness and recoverability as strategies to improve the fault tolerance of the system. However, the disruptions continue to appear in the railway network causing delays, cancellations of trains, etc.

A palliative strategy is to make up for passengers, who suffer these adverse situations. A set of commercial actions can be provided free of charges to these passengers in order to improve their satisfaction.

We propose an expert system to recommend these commercial actions to the passengers. The system, which can belong to a CRM (Customer Relationship Management), consists of two stages. At the first stage, a taxonomy of the passengers is built on the basis of two KPIs (Key Performance Indicators). The KPI1 is the value of the customer for the railway operator and the KPI2 is the satisfaction of the passenger. Both KPIs define a set of patterns of users. A multiobjective linear approach allows to build a master plan of the distribution of commercial actions among the set of patterns of passengers. The second stage consists of the assignment of these commercial actions to the individuals in real-time. If a passenger buys a ticket via an online sales channel then the system detects the features of this customer. A set of rules, the master plan and the current inventory of actions allow the expert system to provide a specific commercial action to this passenger if it is appropriate.

ABSTRACT. Meeting the needs of passengers will increasingly become a competitive factor for airports. Information is one of the most valued services for passengers. Therefore, timely providing of real-time data to the passengers is high on the list of airlines and airports. Since not every passenger has a mobile device at hand, relevant information need also be disseminated offline. Static and dynamic signs cover the above mentioned issues, but the place, where such signs are installed is crucial. 

The goal of our new design concept for terminals is providing information to passengers, where and when they need it. Each service point --- from check-in through security check to boarding --- has been designed using curved layout design and furniture along with large displays for information. The size of the displays is chosen such that passengers are able to recognise important information with respect to the specific service point at one glance. Waiting seats are surrounding the service point to provide an unhindered view to the displays and to promote communication.

We implemented our design ideas in our artificial terminal building of an international airport. The impact of information displays is modeled by a microscopic simulation. Adopting the assumption that the information displays make it easier for the passengers to figure out the way through the terminal we simulate a whole day at the artificial terminal.

ABSTRACT. Since air traffic volume increased over the oceanic airspaces, it has been primordial to improve oceanic air traffic management procedures. One of the most important limitations in the oceans air traffic is the lack of radar coverage. The availability of new surveillance means, called automated dependence surveillance-broadcast system (ADS-B), permits to enhance the strategic flight planning over the oceans by reducing the separation standards. Besides, oceanic flights are mainly subjected to strong winds caused by the jet streams. In this work, we focus on optimizing the strategic flight planning over the North Atlantic airspace. First, we organize the traffic inside a route structure that benefits from both the Jet streams and the exploitation of ADS-B systems. Indeed, from one side, these routes are merged inside the jet streams in order to be as close as possible from wind-optimal routes. On the other side, these routes are constructed to fit in with the new separation standards required when implementing the ADS-B systems. Then, we resolve conflicts between aircraft via an optimization model based on a speed regulation. 
Simulations were conducted for a real traffic data. Computational findings show that the proposed methodology provides satisfying results.