ABSTRACT. The Keynote Lecture has an instructional character. It presents basic concepts and ideas of intelligent decision making for transportation and logistics. The application of different modern technologies and methodologies to generate good and rational decisions is discussed. The State of the Art in the Methodology of Multiple Criteria Decision Making / Aiding is presented and its extension towards Intelligent Decision Making is demonstrated. The "intelligence" in different phases of the decision making process, including: problem recognition, modeling and solving is presented. Several transportation-logistics case studies show the applicability of the concept of "Intelligent Decision Making" in these areas.

ABSTRACT. The tourism sector has been facing continuous growth. It plays a vital role in the economic development of countries, highlighting the need to keep nurturing it by making it easier and more attractive. This paper presents Tourism as a Service (TaaS) as an innovative concept that aims to ease a day in the life of a tourist, integrating services that can be found spread over separate tools and services, including ticketing in public transport and touristic attractions, route planning, information, among others. Its conceptualization and design followed a user-centered design approach. First, focus groups were conducted to understand the needs of users regarding the use of a mobile-based solution in the context of tourism. The findings of the literature reviewed and of the previous stage were prioritized in terms of relevance in a questionnaire sent to potential users, allowing the creation of a medium-fidelity prototype. The validation through usability tests confirmed the interest in the proposed solution.

ABSTRACT. Shared mobility services are slowly penetrating European cities. Hence, transport models that are capable of modelling them are a necessity, to support the policy-makers for making informed decisions. Given that many European cities, especially small and medium sized ones, continue to use the traditional four step modelling approach and such an approach does not have the necessary capacity to model the shared mobility services, there is a need to extend them. Hence, this research proposes an extended framework, by the addition of modules for synthetic population generation and fleet management. Furthermore, modules are suggested for estimation of emissions, car-ownership and induced demand, as such measures are increasingly expected by cities. Multiple equilibrium checks between the aforementioned modules are avoided in the design of this development, to reduce model complexity and convergence issues. This extended framework provides an opportunity to cities to evaluate and integrate shared mobility systems, and form long term planning strategies.

ABSTRACT. For current and future automated driving functions, the radio availability of broadband hybrid networking services (e.g. digital broadcasting, mobile radio, dedicated short range communication) is a prerequisite for continuous V2X information exchange. The supply focus for this is explicitly the road route with its lanes. The application of antenna-based solutions for such longitudinal radio cells with hybrid telematics services is expensive from the installation point of view and can only be adapted to new future telematics standards with great effort. A more suitable solution for such longitudinally shaped radio cells for road routes are leaky coaxial cables (LCX), which are already successfully used for tunnel solutions, for example. The paper discusses the installation and radio implementation of broadband LCX solutions (up to 6 GHz) in terms of simulation and surveying. The integration of the LCX into the crash barrier is favored due to low installation effort and easy upgradeability. An installation was realized on an automotive test fields, where preliminary empirical results for radio simulation and coverage were obtained. Based on the simulations and evaluation measurements, it can be shown that the propagated coverage approach is sustainable over all radiated services. Further solution approaches such as the direct insertion of LCX into the roadway and the derivation of vehicle location information are discussed in the outlook of the paper.

ABSTRACT. The fast-changing behaviour of people in metropolitan areas is creating several challenges to local authorities in managing the urban space. These changes are strongly related to the evolution of technology and its adoption by companies and citizens. Current regulations need to be rapidly updated to respond to the new urban dynamics. However, the gap between local authorities and citizens and the communication difficulties are getting bigger as urban centres grow, creating obstacles to innovation and hindering the deployment of new mobility solutions. The low levels of participation in public consultation actions decrease the quality of new policies, as well as their acceptance by the community. Not only do cities need to be reinvented, but local authorities also need to rethink how to interact with citizens, competing for attention in a digital world. Although digital tools are of easy access, they are not available to everyone, and municipalities need to consider both digital and non-digital interactions to assure that all citizens are able to participate. In this work, we analyse and compare a set of measures that municipalities have been adopting to increase citizens’ engagement, and develop a methodology to help local authorities increase public participation and improve citizens’ commitment towards the city.

ABSTRACT. This paper investigates the characteristics of passengers' behavior and its effect on the metro stopping time at stations. A field survey was conducted on two mega stations in Cairo metro by video cameras photographing and a stated preference survey was performed based on design of experiments approach. The level of passenger non-compliance behavior, defined as the passengers start boarding the metro while others alighting is analyzed and its relationship with deferent affecting factors is fitted by a two-factor interaction model with a determination coefficient of 0.9234. It is found that some factors have a significant impact on passenger’s non-compliance behavior such as the number of alighting and boarding passengers, applying a penalty when not comply with the rules and the door opening shift (i.e. the train stop away from the place designated for it on the platform). Factor interactions show that alighting process does not affect passenger behavior in case of door opening shift. This may be because the alighting passengers are ready to alight from their position inside the train and have nothing to do. Delay from work is the most influential factor and has interaction with all affecting factors. Results may help for verifying passenger behavior simulation to improve the operational capacity of the metro network.

ABSTRACT. Train delays are often monitored using lag indicators of achieved punctuality levels, but indicators can also be used proactively as lead indicators to help reach a certain level of punctuality. In this paper, two lead indicators – share of extended dwell times and share of extended run times – are formulated and tested on Swedish commuter train data for the years 2010-2018. Results indicate that no more than 13 per cent of dwell times and 6 per cent of run times can be extended to reach 95 per cent overall punctuality. Furthermore, the results show that none of the three commuter train regions in Sweden reach the target level for extended dwell times and only one of them for extended run times.

ABSTRACT. In railway traffic, perturbations may give rise to conflicts, causing delays. As a countermeasure, effective re-scheduling and re-routing decisions can be taken by addressing the real-time Rail Traffic Management Problem (rtRTMP). One of its subproblems is the real-time Energy Consumption Minimization Problem (rtECMP). The latter enforces the train routing and precedences computed by a rtRTMP solver and defines train timings and speed profiles. The objective is to minimize the weighted sum of train energy consumption and total delay. In this paper, we propose an Ant Colony Optimization algorithm for the rtECMP and we test it on the French Pierrefitte-Gonesse control area with dense mixed traffic. The results show that, in a very short computing time, a remarkable exploration of the search space is performed before convergence.

ABSTRACT. Rail operations are usually affected by incidents that may cause traffic to deviate from the planned operations making impossible to operate the schedule as it was planned. In such a situation the operator needs to adjust the schedule to get back to the original schedule. A train operator may have the policy of economically compensating (e.g., refunding ticket fare) passengers when they incur in delays. Compensation levels usually depend in the amount of delay. Therefore, it is important to have a smart way of deciding whether to speed up trains in order to absorb delays, i.e., increasing energy consumption, or to compensate passengers. In this paper a mathematical model which decides on the speed profile while considering passenger use is presented. The model decides on the optimal sequence of operating regimes and the switching points between them for a range of different circumstances and train types all while considering delays and passenger compensation policies. The objective is to minimise both energy consumed and incurred compensation to passengers. Constraints on traction and braking forces,on train velocity, on forces caused by vertical and horizontal track profile, and on passenger compensation policy are considered. Computational tests on realistic problem instances of the Spanish rail operator RENFE are reported. The proposed approach is able to find solutions with a very good balance between various managerial goals.

ABSTRACT. Cycling and micro-mobility, in general, have been long promoted as sustainable and favorable modes of transport due to emission mitigation, congestion reduction and improvements to users’ health and lifestyle. However, as most cities in the world have followed a car-centric development, their bicycle network is often highly fragmented, constituting the biggest barrier for attracting new users. This paper introduces a data-driven procedure based on micro-mobility geo-referenced data collected in the city of Rome (Italy). The aim is to identify corridors of high-density micro-mobility demand through an iterative clustering procedure and to evaluate potential growth scenarios of the bicycle network by locating the strategic missing links in the existing infrastructure to achieve a fully connected bicycle network that maximizes the overall usage of deployed bike lanes.

ABSTRACT. During the past decades, unlike other countries of the European community, Greek society excluded bicycle from its everyday life, both as a solution for sports and recreation, as well as a transport mode. However, in the last few years, maybe due to the economic crisis, people in Greece began to consider cycling, again, in their daily routine. In this context, the present paper endeavours to examine and evaluate the ability to cycle in the complex urban environment of Greek cities. Therefore, this study develops a method for assessing bikeability, based on international literature and related walkability evaluation methods. Initially, a literature review was conducted to identify those parameters that positively affect the suitability of a cycling route. Afterwards, we developed linear scales to transform the quantitative data into scores ranging from 0 to 10, in order to be comparable. In the next step we used participative Analytic Hierarchy Process (AHP) in order to weight each parameter and identify its significance. Lastly, we formulated the final bikeability index. The chosen case study is the Municipality of Zografou, which constitutes one of the most significant municipalities in Athens, combining all aspects of a modern city: a liveable central area, university campuses, high residential density and traffic congestion problems. The results indicate that the overall bikeabilty index of Zografou is quite low, specifically it equals to 3,7/10 meaning that most of the road sections are not suitable or attractive for cycling. This paper provides fruitful insights about the issue of bikeability in urban areas and mostly its spatial dimension and it could be used as decision making tool for stakeholders and planners.

ABSTRACT. Promoting pedestrian mobility represents a strategy to achieve a sustainable transportation system, where problems such as traffic congestion, air and noise pollution are minimized. For this aim, it becomes fundamental to identify pedestrian environments that can facilitate and encourage to make trips by walking. The objective of this paper is to provide a practical methodology that can assist the analyst to identify the best alternative among some pedestrian paths with different characteristics. To this aim, subjective, objective and mixed indicators were proposed to measure the performance criteria adopted for evaluating the alternatives. A Multi-criteria analysis was applied as a tool of evaluation. The findings of the work suggest that pedestrians seem to appreciate the tree-lined paths, and prefer to keep bike lines separated from pedestrian paths. Number of crosswalk is relevant in the evaluation of pedestrian path, especially when there are shops along the path. Aspects related to the conditions of the pavement and to comfort and environment are less relevant.

ABSTRACT. Bicycle sharing systems (BSS) have the potential to contribute to the creation of a cycling culture, by normalizing cycling and providing access to bicycles. This research looks at the use of BSS in ‘starter’ cycling cities, where modal share of cycling thus far is low and there is only limited cycling infrastructure. Surveys with users of the BSS in Limassol (Cyprus), Las Palmas de Gran Canaria and Malta shed light on “who” uses the BSS and “why”. Through descriptive statistics and binary logistic regression models, the influence of individual, social environment and physical environment factors on shared bicycle use is analysed, looking at differences between frequent and infrequent BSS users, to get a better understanding of the motivators and barriers that influence BSS use. Frequent BSS use is positively associated with frequent use of other ‘alternative’ transport modes, such as walking and public transport use, as well as with shorter distances from respondents’ residence and most frequent destinations to the nearest BSS station. Higher perceived safety of cycling was also associated with more frequent BSS use, as did a positive social norm, including support from friends and family, respect from other road users, and feeling that cycling is an accepted form of transport, confirming the importance of such factors in building a cycling culture.

ABSTRACT. When to use the mesoscopic level when modelling traffic?

Porto Municipality and PTV will be co-presenting the case of Porto at the EWGT. A mesoscopic model using Intersection Capacity Assignment to calculate intersection capacity in the detail, without losing the overall network analysis capability.

Presenters:

Cristina Vilarinho, Camara Municipal de Porto

Ignacio Galindo Pinto, PTV Group Iberia

ABSTRACT. Multicollinearity is a statistical phenomenon in which predictor variables of a multiple regression model are highly correlated; this may affect model reliability and interpretability. It is a common problem for models with a large number of predictors, as it often is the case in real-time road safety analysis models. This work introduces the principal component logistic regression (PCLR) model to this field, in order to deal with the issue of multicollinearity. A real-time conflict prediction model to predict rear-end conflicts was calibrated with both a classic logistic regression (LR) and PCLR. Data were collected on an Italian motorway for one year, with radar sensor on multiple cross-sections. According to our findings, LR coefficient estimates were not reliable, as several of them were not coherent with observed descriptive statistic values and with literature findings. Conversely, PCLR provided significant improvements in terms of model interpretability, leading to a model which is able to predict unsafe situations with a satisfactory precision, and to suggest on which variables to intervene in order to avoid such situations.

ABSTRACT. The main objective of the geometric design consistency is to minimize the emergence of unforeseen events when road users drive along a road segment. A consistent highway design ensures that successive geometric elements act in a coordinated way. In India, vehicles with diverse static and dynamic characteristics interact with each other leading to a significant difference in traffic flow operations compared to highways in developed countries. Due to the heterogeneity in traffic flow, geometric consistency would vary significantly by vehicle category for similar curve characteristics. The present study analyzes geometric design consistency by vehicle and develops vehicle category-based design consistency models for mixed traffic environments on two-lane rural highways in hilly terrain in India. A total of 30 curves with varying geometric characteristics located along the National Highway (NH-953) connecting Netrang and Rajpipla, Gujarat, India, were selected. Speed data were collected using radar gun for three vehicles category: motorized two-wheelers (MTW), cars, and heavy commercial vehicles (HCV). Geometric design consistency was evaluated using operating speed as a surrogate safety measure. The results showed that 38% and 10% of the curves (for cars), 51% and 20% of the curves (for MTW), and 32% and 51% of the curves (for HCV) have ratings of fair and poor consistency, respectively. Design consistency models for different vehicle categories were developed using the generalized linear regression modelling (GzLM) technique. The results showed that the deflection angle is positively correlated, whereas curve and tangent lengths are negatively correlated with highway alignment design consistency.

ABSTRACT. An intersection is a critical component of a given road network, where traffic from different approaches merges and diverges. These merging and diverging traffic movements have significant impacts on traffic operations and safety, particularly at multiple-approach intersections. Different traffic management measures and control strategies, along-with necessary geometric improvements, are suggested to optimize traffic operation at a given intersection. It is also more prudent to assess the effectiveness of potential strategies towards improvement using simulation first; before certain appropriate strategies/alternatives are finally selected for real field implementation. Kamrej intersection, a multi-legged junction in Surat, located in the western part of India, is taken as a case study. Traffic data obtained from real field conditions is used to calibrate the base model (present-day status of intersection) using VISSIM and replicate actual field conditions. The calibrated model is then validated by means of vehicle-class-wise average travel time for each of the traffic movements (straight and turning) and traffic flow during peak hours. Average delay is used as an additional measure to assess the effectiveness of selected potential traffic management alternatives. When different alternatives are simulated, it is found that delays for straight and right movements can be reduced substantially, contributing towards improvement in efficiency and traffic operations at selected intersections. The present approach holds promise, thereby contributes to improving traffic operations at selected multi-legged intersections.

ABSTRACT. E-commerce is a continuously growing sector worldwide, with important repercussions on the delivery system in urban areas and especially in the Business to Consumer (B2C) sector. The delivery of a package to a consumer's address involves not only high costs for couriers (greater number of kilometres travelled), but also increased congestion and greater environmental pollution (greater volume of pollutants released into the air). To rationalize deliveries in urban areas the use of collection points, equipped with lockers, to store the goods that users have ordered has been considered in literature. This work compares two alternative delivery options: deliveries to the consumer's home versus to Lockers. To make this comparison we used a cluster first route second math-heuristic approach. In the clustering phase, we experimented a new clustering function, while the routing phase consists in solving an instance of the Traveling Salesman Problem for each generated cluster. Finally, we applied the math-heuristic to a real case (the Italian municipality of Dolo near Venice) and compared the two delivery alternatives. We evaluate the performance considering two different fleets of vehicles, with small and medium capacity. In addition, since additional trips might be performed by consumers to pick up parcels at Lockers, a sensitivity analysis was carried out to analyse the sustainability of the proposed city logistics scheme.

ABSTRACT. Social distancing guidelines established amid the COVID-19 pandemic have decreased the generation of trips in urban transportation networks; furthermore, travelers have shifted away from high occupancy modes due to the fear of contagion. This scenario has led to reduced public transportation ridership and increased shares of private motoring, walking and cycling trips in urban areas. Reduced fare collection, in addition to occupancy restrictions, have imposed unprecedented challenges for transit operators; on the other hand, policy makers have found a unique opportunity to stimulate more sustainable travel habits. In Budapest, preliminary data from March 2020 suggests that stay-at-home orders have resulted in intra-city and commuter traffic reductions of over 30%, while walking and biking public transportation ridership decreased as much as 90% in the same period. This scenario suggests that the COVID-19 pandemic has driven significant modal share changes in Budapest and poses uncertainty on traveler behavior in the post-lockdown phase. This article aims to further investigate the effects of the COVID-19 pandemic in the demand for urban transportation in Budapest and to simulate them in a macroscopic model of the city.

ABSTRACT. COVID-19 pandemic has a significant influence on people's lifestyles including their travel choices. Due to the pan-demic, restrictions in travelling have taken place throughout Italy because there was an obligatory need for social distancing and quotas on public transport. Road users in the city started to use their private vehicles more for long distances, and consider walking and cycling for short-distance trips. Governments and local authorities encouraged citizens to use sustainable travel modes, particularly walking, during the pandemic period. However, high number of infections, in especially countries which had high number of death, have strongly influenced the propensity of walk-ing due to the psycho-social aspects of travelling. This paper presented a statistical analysis that summarised and interpreted the data gathered through questionnaires in the urban areas of Sicily. This study was focused on travelling by walking for either leisure or work. The analysis focused on an evaluation of negative emotions experienced by a sample of road users who habitually have been walking for short distances in the area of study. The analysis was conducted in a bivariate manner. The data indicated a variation between three emotions; anxiety, stress and fear. These emotions had a potential to influence people’s daily life and, as a result, their travel habits.

ABSTRACT. The Covid-19 crisis forced governments around the world to rapidly enact several restrictions to face the associated health emergency. The Portuguese government was no exception and, following the example of other countries, established various limitations to flat the contagions curve. This led to inevitable repercussions on mobility and environmental indicators including noise. This research aims to assess the impact of the lockdown due to Covid-19 disease on the noise levels recorded in the city of Porto, Portugal. Data from four noise sensors located in strategic spots of the city were used to calibrate and validate Time Series Models, allowing to impute the missing values in the datasets and rebuild them. The trend and the cyclic information were extracted from the reconstructed datasets using decomposition techniques. Finally, a Spearman correlation analysis between noise levels values and traffic volumes (extracted from five inductive loop detectors, located nearby the noise sensors) was performed. Results show that the noise levels series present a daily seasonal pattern and the trends values decreased from 6.7 to 7.5 dBA during the first lockdown period (March-May 2020). Moreover, the noise levels tend to gradually rise after the removal of restrictions. Finally, there is a monotonic relationship between noise levels and traffic volumes values, as confirmed by the positive moderate-to-high correlation coefficients found, and the sharp drop of the former during March-May 2020 can be attributed to the strong reduction of road traffic flows in the city.

ABSTRACT. Public transport services play an important role in the mobility of the population in urban centers, allowing a decrease in the number of private vehicles in circulation and contributing to a more sustainable mobility. However, the emergence of the COVID-19 pandemic had a serious impact on the mobility habits of the population, with a substantial reduction in the number of public transport passengers due to the fear of contagion, which raises questions about the future sustainability of cities. Thus, it is essential to restore the confidence of travelers to feel safe and comfortable using public transport services. Taking advantage of the widespread use of mobile technologies, this article intends to propose a route planning system for public transport that meets the needs of passengers in terms of safety and comfort. After a systematic review of the existing literature and a series of focus group sessions, a prototype of the system was developed, and subsequently evaluated by potential users through usability tests. The results obtained are a good indicator of the system's functionality and ease of use. This assessment allowed us to corroborate the potential that the proposed route planning system has in promoting the use of public transport services as a means of mobility.

ABSTRACT. The Vehicle Routing Problem with Time Windows (VRPTW) asks for the optimal set of routes to be performed by a fleet of vehicles to serve a set of customers within their assigned time windows. In this work, we propose a matheuristic for the VRPTW which utilizes the sub-problem constituted by optimizing only a selected time window of the VRPTW whereas the tours outside of this time window are regarded as fixed. We call this problem the Single Time Window Vehicle Routing Problem (STWVRP). For applying the STWVRP, we must assume that several customers are assigned to the same time window, i.e., the number of time windows is much smaller than the number of customers. An exact problem description of the STWVRP is given in the form of a Mixed-Integer Linear Programming formulation. We apply this exact formulation within a matheuristic for the VRPTW. The paper concludes with extensive computational experiments.

ABSTRACT. New technological innovations of eco-friendly vehicles for freight transport combined with the usage of renewable energy sources showed significant results in mitigating transport-related carbon footprint. Therefore, in this paper, we present a multi-objective network design model considering a novel technology, the eHighway system, based on electrified roads to supply new Overhead Catenary (OC) hybrid trucks. This work investigates the opportunities of adopting eHighways and evaluates its environmental benefits considering limited budget resources for infrastructure electrification. We propose an optimization problem formulation including three objectives: the minimization of infrastructure and environmental costs, and the maximization of the total number of OC hybrid trucks served on electrified arcs. The Pareto optimization approach is considered for a comprehensive analysis of all possible solutions according to different criteria weights. The proposed model has been evaluated on a test network and the numerical results of Pareto optimization show the environmental improvement we can obtain by using the eHighway system up to about 99% according to the assumed available budget and assigned criteria weights. As a result, the model can be considered as a useful tool for decision-makers in the eHighway network design.

ABSTRACT. The impact of travel time unreliability on freight transport has been extensively investigated in the last 20 years with the main focus being placed on mode and route choice (and using mostly stated preference data). In contrast, freight distribution has been very rarely examined. In this paper, we describe a study on how travel time unreliability affects interregional freight distribution using (revealed preference) data from the last national transport survey carried out in Iran (2015). Through this study, conducted using spatial interaction linear and geographically-weighted regression models, we found that, globally, travel time reliability is approximately as important as average travel time in determining freight distribution flows, but this importance varies widely across regions. We also found that tardy trip reliability measures describe freight distribution patterns more accurately than statistical range measures (coefficient of variation of travel time).

ABSTRACT. This paper proposes a new agent-based model (ABM) to explore different scenarios of e-commerce urban deliveries, comparing door-to-door deliveries with consolidation-based strategies. The ABM reproduces operation under different demand patterns and include the possible matching of customer systematic trips and collection/delivery points with small detour from the scheduled trip. Several variables of the model can be changed in a parametric simulation environment, allowing to infer the level of convenience of consolidation strategies for the different actors involved. The model provides indicators able to take into account customer and logistics operator perspectives, and the impact of the service on the community. Results can give useful information to understand how to manage growing on-demand urban deliveries and to measure the impact of freight transport on city sustainability.

ABSTRACT. The era of intelligent transportation with automated vehicles (AVs) is coming. Nonetheless, the transition to this system will be a gradual process. On the one hand, some zones in the city may be dedicated to AVs with a fully intelligent traffic management system geared toward high performance. On the other hand, automated and conventional vehicles may have to be allowed to drive in the remaining zones of the urban network in a transition stage. In this paper, we consider a situation where AVs are deployed by a taxi operating company to serve door-to-door travel requests. Facing this transition period, a strategic flow-based vehicle routing model is developed to determine the optimal fleet size of automated and conventional taxis as a function of the gradually increasing coverage of the AVs-only dedicated area. Traffic congestion is considered through flow-dependent travel times. Two taxi company service regimes are tested: the User Preference Mode (UPM) and the System Profit Mode (SPM). In the UPM, passengers can choose their preferred vehicle type according to their preference. In the SPM, the taxi company will take charge of the vehicle assignment to maximize the system profit. The developed model formulations are applied to a case study of a large toy network. The results give insight into the performance of the heterogeneous taxi system on a hybrid network. Strategies are presented on how to adjust the fleet size of automated and conventional taxis to get the best system profit while satisfying the mobility demand. The SPM can bring more profit to the operating company by reducing the detour and relocation cost of taxis, reducing the salaries for drivers through a bigger fleet size of AVs, and reducing the delay penalty, compared to the UPM.

ABSTRACT. Growing development in technologies that can lead to fully automated driving is at pace. This can result in an enormous change in traffic operations and network properties. However, there are uncertainties about the full deployment time of these automated vehicles on road networks. The transition period from vehicles with drivers to the driver-less will result in a mutual environment with an interaction between traditional (that is, manual) vehicles, automated vehicles and infrastructure. In this context, this research focuses on the various factors of land use, user demographics and road network affecting the percentage of automated vehicles into the multi-vehicle assignment models and their subsequent impacts on the traffic network properties. This research aims to use a realistic approach to input the percentage of automated vehicles into the traffic models via an indicator matrix, and to evaluate the impacts by applying it on a real-world network through stochastic user equilibrium assignment.

ABSTRACT. In this article, the basic Reinforcement Learning (RL) concepts are discussed, continued with a brief explanation of Markov Decision Processes (MDPs). The reasoning for the application of RL in the autonomous vehicle control domain is accompanied by a developed basic environment for simulation-based training of agents. Furthermore, we look at the available literature of successor features, and the recent achievements of its utilization. After explaining its methodology and showing how it works, state-of-the-art ideas are investigated and infused into the vehicle control realm. Moreover, the paper details how these features can be tailored to the Highway driving domain and what is the secret behind its capability to boost the performance of the RL agent. The results imply that learned skills can help with the multi-objective rewarding problem, and agents applied to changing reward systems can adapt quickly to the new tasks. The only thing to find is the correct decomposition and selection of successor features.

ABSTRACT. Autonomous delivery devices (ADDs) are medium-size autonomous vehicles used for urban logistics operations. They are expected to highly improve the efficiency of last-mile distribution in dense urban environments in future years. However, the magnitude of the operation cost reduction greatly depends on some operation variables of the robots. The main objective of this paper is to quantify the ADD operation cost uncertainty depending on the stochastic behavior of given input variables. First, ADD operations are modelled using the continuous approximation methodology. These mathematical formulations relate some ADD operation input parameters – mainly the volume capacity of the robot, its commercial speed, its unit operations costs, and its maximum range – and give an estimation of the carrier’s last-mile operation costs. Then, an uncertainty analysis based on the Monte Carlo approach is done. To the best of our knowledge, this is completely novel in the field of ADD operations assessment. Instead of working with deterministic values, some probability distribution functions (PDFs) are assumed for the different input parameters. The ADD total operation costs consequently follow their own PDF that is obtained via the Monte Carlo process. The model uncertainty analysis indicates that ADDs would, on average, reduce last-mile operation costs by 8% for a demand density around 40 receivers/km2.

ABSTRACT. This work proposes an architecture to treat georeferenced data from the OpenStreetMap to plan routes. The methodology considers the following steps: collecting data, incorporating data into a data manager, importing data into a data model, executing routing algorithms, and visualizing routes. Our proposal incorporates the following features characterizing each street segment: safety &security, comfort, accessibility, air quality, time, and distance. Routes can be calculated considering any specified weighting system of these features. The outcome of the application of this architecture allows to calculate and visualize routes from georeferenced data, which can support researchers in the study of multi-criteria routes. Furthermore, this framework enhances the OSM data model adding a multi-criteria dimension for route planning.

ABSTRACT. The Dial-a-Ride Problem (DARP) consists of designing a schedule for a set of customer requests served by a fleet of vehicles. Further restrictions, namely time windows, pick-up and drop-off locations and capacity constraints are generally customary. A common example arises in door-to-door transportation of people, especially elderly people or people with impaired mobility. In this work, we consider a dynamic-deterministic variant of the DARP for two Austrian mobility providers. The focus of these operators lies on rural regions served by a heterogeneous fleet of vehicles. As additional conditions we take the transportation of people with impaired mobility, break times and the capacity of wheelchairs into account. We consider a heuristic and an exact solution approach for the DARP, namely a Large Neighborhood Search and a Mixed-Integer Linear Programming approach. In a computational study, we show different operator scenarios (minimizing overall driven kilometers, number of used vehicles, and number of unscheduled requests) with up to 500 requests per day and 30 vehicles.

ABSTRACT. Bike sharing is one of the recent trends in urban areas as a solution for the mobility problem, where it is seen as a significant component of local transportation due to the need to ensure last-mile transportation. Bike sharing systems enable bicycles to be rented at a station for a ride and be dropped off at another station which provides flexibility to users. However, incorporating a bike sharing system into the transportation network of a city involves its own challenges including balancing the supply and demand within stations, and scheduling bikes’ repositioning. The bike repositioning scheduling problem arises due to the variable demand for various types of bicycles that can be rented from and left to any station. In order to ensure sustainability in bike sharing operations, bikes should be redistributed to the stations considering the demand, during for example a day or night. In this context, with the ultimate motivation of proposing a bike sharing system to meet within-day demand dynamics in a metropolitan city, we address the Static Bike sharing Rebalancing Problem (SBRP) with multiple capacitated vehicles for redistribution of bicycles. On purpose, we formulate a mixed-integer linear program in which the objective is minimizing the total cost subject to a set of constraints including truck capacity, demand satisfaction, inventory balance, and flow preservation. We discuss the results from solutions to the SBRP through a real case study.

ABSTRACT. The usage of carsharing systems is increasing in cities world-wide. Allowing one-way trips to extend the system’s flexibility causes vehicle imbalances across stations, which can be solved by relocating cars. Various research has been conducted on relocation planning, both for conventional and autonomous vehicles. We introduce a novel repositioning approach that is intended for partly autonomous vehicles, based on the upcoming technology of platooning. We propose an integer program and solve several instances to depict advantages and limitations of the platooning approach. We further compare the new approach to conventional repositioning in terms of saving costs and realizing more customer trips.

ABSTRACT. There is no data collected and saved about the road signs in Sweden and the status for these signs is unknown. The data used in this paper were collected in a previous work [1] done by the Swedish National Road and Transport Research Institute (VTI). The status of colours, quality of retroreflection, type, and age of the road signs are unknown. The aim of this study is to use the supervised machine learning to predict the status and quality of the road signs. The status of a road sign (approved or not) depends on the retroreflectivity and the colours. This study even investigates the effect of using principal component analysis (PCA) and data scaling on the accuracy of the prediction. The data were prepared before using then scaled using two methods (Min-Max Scaler and StandardScaler). Using PCA to minimize the dimensions of the data improved the accuracy of the predicting only when the data unscaled. The PCA gives lower predicting accuracy when using scaling compared with accuracy when using unscaled data. When no PCA used, scaling the data improves the accuracy of prediction compared with unscaled data. The scaled data using StandardScaler (data standardization) gives the most accurate predictions. The learning models used in this study gives a very good predicting accuracy and can be used for predicting the status of the road signs mounted in the roads.

ABSTRACT. Intelligent intersection management systems are an integral part of Smart Cities and have a profound impact in urban traffic management. In a previous work, the authors proposed a specific Intelligent Intersection Management Architecture (IIMA) with the associated Synchronous Intersection Management Protocol (SIMP) for simple single-lane isolated intersections that outperformed other competing protocols in throughput, time loss and polluting emissions. IIMA/SIMP supports both autonomous and human-driven vehicles. This paper extends such work to more complex multi-lane intersections, comparing against traditional and intelligent intersection management approaches. Simulation results achieved with SUMO confirm the advantages of IIMA/SIMP even in complex intersections, improving throughput, average speed, waiting time, trip time loss, and associated fuel consumption.

ABSTRACT. Several tradable credit schemes have been proposed over the last decade to restrict the use of personal cars and reduce negative traffic externalities such as congestion and pollution. Two of the main arguments of this approach compared to congestion pricing are that it is revenue-neutral and that the market could self-regulate the credit price. The trip-based MFD framework is an efficient tool to simulate traffic dynamics at a large-urban scale considering multimodal options. It is very convenient to test demand management strategies targeting all travelers. In this paper, we propose to use such a simulation framework to investigate a tradable credit scheme, which aims to foster mode shift by providing limited access to the road network. Credits are allocated to the users, and they need to pay a toll with credits when using their car. They can save their credits for another day or trade them using a marketplace. The credit scheme impacts are illustrated through a numerical implementation with a scaled demand typical of a peak hour (7:00 to 8:00) in Lyon with 1177 trips. The main result of the simulation is the potential for such a tradable credit scheme to increase PT share and thus reduce total travel costs.

ABSTRACT. As cities across the globe continue to grow, traffic congestion has become globally ubiquitous with great economic and environmental costs associated with it. The increasing prevalence of self-driving vehicles creates an opportunity to build smart, responsive traffic infrastructure of the future. Such an infrastructure consisting of connected and autonomous vehicles and smart traffic lights would have the potential to cope with congestion, weather phenomena and accidents, while maintaining safety and ensuring privacy of information. This paper introduces an approach to address the challenge of dynamically adjusting traffic to the changes in the environment. We argue that multiagent meta-level control (MMLC) is an effective way to non-myopically determine how and when this adaptation should be done. The approach highlights the role of dynamic meta-reasoning in a platooning scenario, in which collaboration contributes to improved travel time for vehicles in the network as well as a positive environmental impact as related to fuel consumption and emissions. Specifically, for the case study described in the paper, our MMLC-based approach leads to approximately 44% decrease in travel time, 7% increase in average speed, a 32% decrease in fuel consumption and a 35% drop in emissions. We also see performance advantages for a scaled-up mixed traffic simulation environment.

ABSTRACT. Collecting travel data in the field is always a challenging task. It's equally burdensome to respondents if the data is collected using face-to-face personal interview or self-completion surveys. To reduce the burden on the respondent and to collect the time stamps and locations precisely, a few fully automated survey approaches are proposed with limited success mainly if the required sample rate is higher in a large urban agglomeration. This study presents an open-source, web-based, self-completion and/ or personal-interview survey platform, namely Travel Survey as a Service (TSaaS) which currently hosts three different survey types. This study proposes to use the TSaaS platform as the crowd-sourced data collection approach for household travel diaries. A pilot study was conducted in Jaipur, and three different data collection approaches are attempted. The approaches are compared in terms of survey completion rate, survey completion time and time-cost of each approach. The crowd-sourced web-survey turn out to be the most efficient in terms of the time-cost per completed survey record and most suitable to collect a large number of survey records in an urban agglomeration.

ABSTRACT. Dynamic mode choice is essential to understand the potential effectiveness of policies aiming to achieve desirable modal split targets or to manage the demand for resource-limited systems such as shared mobility services. In this paper, we propose an estimation of dynamic modal split for work-related trips, including mode- and time-specific costs, with activity participation based on utility maximization. In order to obtain an accurate profile while remaining at an aggregate level, three types of work activities are described (full time, morning and afternoon shift). The estimated modal split concerns motorized vehicles, soft modes but also train and urban public transport. Based on utility maximization principles, the accumulated utility is formulated within a departure time choice model. A Markov Chain Monte Carlo procedure is used to evaluate the marginal utility function parameters which are used in a joint departure time and mode choice evaluation. Mode specific travel speed for each time of the day is used to estimate also travelled distance distribution per mode. The methodology is applied and tested, using data collected in Ghent in 2008. 16.749 work-related trips have been considered in a simplified estimation where two successive trips are constrained to be done with the same mode. This methodology is characterized by low data requirements and the model is shown to be flexible to include all available type of information in order to refine or accelerate the estimation. The proposed method is easy to implement using only dynamic trip counts, without the need for simulation or traffic assignment.

ABSTRACT. Day-to-day dynamics of mode choice adaptation to a radical supply change, such as the introduction of a new metro line, is considered. A model based on a stochastic process of choice is proposed. The variables of the dynamical system are choice probabilities. The model considers inertia, i.e. state dependence, and shocks, i.e. serially uncorrelated random terms, and is Markov. The system evolves according to time-homogeneous transition probabilities. A comprehensive review of the properties related to convergence to the stationary state and speed of convergence is provided. The problem of estimating the entries of the matrix of transition probabilities when these are logit with inertia is addressed. The insights are illustrated by the analysis of the dynamics, from a stationary state to a new stationary state, in a two-alternative numerical example based on simulated data. In particular, the application investigates the sensitivity of the dynamics to the inertia variable, and provides an assessment of the ability of the maximum likelihood estimators to recover the true coefficients.

ABSTRACT. The pervasive use of mobile devices has brought a valuable new source of data. The work presented here has a twofold objective: firstly, to demonstrate the capability of mobile phone records to feed traditional trip-based demand models and, secondly, to assert the possibilities of using developed models to estimate the effects of new land use development scenarios. Detailed trip data for the metropolitan area of Barcelona are reconstructed from mobile phone records. This information is then employed as input for building a set of demand trip-based models and to apply these daily-based models to the appraisal of new development scenarios in a VISUM model of the city. The model calibration and validation process proves the quality of the models obtained. Our results show the way in which the generated trips are distributed into the study area and modal share is modified in the considered scenarios.