Showing papers in "Transportation Research Part B-methodological in 2019"

Ridesourcing systems: A framework and review

Abstract: With the rapid development and popularization of mobile and wireless communication technologies, ridesourcing companies have been able to leverage internet-based platforms to operate e-hailing services in many cities around the world. These companies connect passengers and drivers in real time and are disruptively changing the transportation industry. As pioneers in a general sharing economy context, ridesourcing shared transportation platforms consist of a typical two-sided market. On the demand side, passengers are sensitive to the price and quality of the service. On the supply side, drivers, as freelancers, make working decisions flexibly based on their income from the platform and many other factors. Diverse variables and factors in the system are strongly endogenous and interactively dependent. How to design and operate ridesourcing systems is vital—and challenging—for all stakeholders: passengers/users, drivers/service providers, platforms, policy makers, and the general public. In this paper, we propose a general framework to describe ridesourcing systems. This framework can aid understanding of the interactions between endogenous and exogenous variables, their changes in response to platforms’ operational strategies and decisions, multiple system objectives, and market equilibria in a dynamic manner. Under the proposed general framework, we summarize important research problems and the corresponding methodologies that have been and are being developed and implemented to address these problems. We conduct a comprehensive review of the literature on these problems in different areas from diverse perspectives, including (1) demand and pricing, (2) supply and incentives, (3) platform operations, and (4) competition, impacts, and regulations. The proposed framework and the review also suggest many avenues requiring future research.

Vehicle routing problem with drones

Abstract: The vehicle routing problem with drones (VRPD) is an extension of the classic capacitated vehicle routing problem, where not only trucks but drones are used to deliver parcels to customers. One distinctive feature of the VRPD is that a drone may travel with a truck, take off from its stop to serve customers, and land at a service hub to travel with another truck as long as the flying range and loading capacity limitations are satisfied. Routing trucks and drones in an integrated manner makes the problem much more challenging and different from classical vehicle routing literature. We propose a mixed integer programming model, and develop a branch-and-price algorithm. Extensive experiments are conducted on the instances randomly generated in a practical setting, and the results demonstrate the good computational performance of the proposed algorithm. We also conduct sensitivity analysis on a key factor that may affect the total cost of a solution.

A survey of models and algorithms for optimizing shared mobility

Abstract: The rise of research into shared mobility systems reflects emerging challenges, such as rising traffic congestion, rising oil prices and rising environmental concern. The operations research community has turned towards more sharable and sustainable systems of transportation. Shared mobility systems can be collapsed into two main streams: Those where people share rides and those where parcel transportation and people transportation are combined. This survey sets out to review recent research in this area, including different optimization approaches, and to provide guidelines and promising directions for future research. It makes a distinction between prearranged and real-time problem settings and their methods of solution, and also gives an overview of real-case applications relevant to the research area.

Literature review of the vehicle relocation problem in one-way car sharing networks

Abstract: In this paper, we perform a systematic review of selected publications that offer method-based solutions to the vehicle relocation issues in car sharing networks. Asymmetric networks allowing one-way trips are the most promising form of car-sharing systems. However, the resulting vehicle imbalance across the station grid requires relocations. Typical approaches to solving the vehicle relocation problem include mixed-integer programming for strategic or operation-oriented design problems, as well as simulation models for management tasks. We survey how researchers define the decision problems related to the vehicle relocation issue, and consider their division into multistage approaches. This article offers a starting point for researchers interested in modeling one-way vehicle sharing.

Distributed model predictive control approach for cooperative car-following with guaranteed local and string stability

Abstract: In this paper, a serial distributed model predictive control (MPC) approach for connected automated vehicles (CAVS) is developed with local stability (disturbance dissipation over time) and multi-criteria string stability (disturbance attenuation through a vehicular string). Two string stability criteria are considered within the proposed MPC: (i) the l∞-norm string stability criterion for attenuation of the maximum disturbance magnitude and (ii) l2-norm string stability criterion for attenuation of disturbance energy. The l∞-norm string stability is achieved by formulating constraints within the MPC based on the future states of the leading CAV, and the l2-norm string stability is achieved by proper weight matrix tuning over a robust positive invariant set. For rigor, mathematical proofs for asymptotical local stability and multi-criteria string stability are provided. Simulation experiments verify that the distributed serial MPC proposed in this study is effective for disturbance attenuation and performs better than traditional MPC without stability guarantee.

The electric vehicle routing problem with energy consumption uncertainty

Abstract: Compared with conventional freight vehicles, electric freight vehicles create less local pollution and are thus generally perceived as a more sustainable means of goods distribution. In urban areas, such vehicles must often perform the entirety of their delivery routes without recharging. However, their energy consumption is subject to a fair amount of uncertainty, which is due to exogenous factors such as the weather and road conditions, endogenous factors such as driver behaviour, and several energy consumption parameters that are difficult to measure precisely. Hence we propose a robust optimization framework to take into account these energy consumption uncertainties in the context of an electric vehicle routing problem. The objective is to determine minimum cost delivery routes capable of providing strong guarantees that a given vehicle will not run out of charge during its route. We formulate the problem as a robust mixed integer linear program and solve small instances to optimality using robust optimization techniques. Furthermore, we develop a two-phase heuristic method based on large neighbourhood search to solve larger instances of the problem, and we conduct several numerical tests to assess the quality of the methodology. The computational experiments illustrate the trade-off between cost and risk, and demonstrate the influence of several parameters on best found solutions. Furthermore, our heuristic identifies 42 new best solutions when tested on instances of the closely related robust capacitated vehicle routing problem.

Optimization models for electric vehicle service operations: A literature review

Abstract: Electric vehicles (EVs) are widely considered to be a solution to the problems of increasing carbon emissions and dependence on fossil fuels. However, the adoption of EVs remains sluggish due to range anxiety, long charging times, and inconvenient and insufficient charging infrastructure. Various problems with EV service operations should be addressed to overcome these challenges. This study reviews the state-of-the-art mathematical modeling-based literature on EV operations management. The literature is classified according to recurring themes, such as EV charging infrastructure planning, EV charging operations, and public policy and business models. In each theme, typical optimization models and algorithms proposed in previous studies are surveyed. The review concludes with a discussion of several possible questions for future research on EV service operations management.

Multiclass traffic assignment model for mixed traffic flow of human-driven vehicles and connected and autonomous vehicles

Abstract: Compared to existing human-driven vehicles (HDVs), connected and autonomous vehicles (CAVs) offer users the potential for reduced value of time, enhanced quality of travel experience, and seamless situational awareness and connectivity. Hence, CAV users can differ in their route choice behavior compared to HDV users, leading to mixed traffic flows that can significantly deviate from the single-class HDV traffic pattern. However, due to the lack of quantitative models, there is limited knowledge on the evolution of mixed traffic flows in a traffic network. To partly bridge this gap, this study proposes a multiclass traffic assignment model, where HDV users and CAV users follow different route choice principles, characterized by the cross-nested logit (CNL) model and user equilibrium (UE) model, respectively. The CNL model captures HDV users’ uncertainty associated with limited knowledge of traffic conditions while overcoming the route overlap issue of logit-based stochastic user equilibrium. The UE model characterizes the CAV's capability for acquiring accurate information on traffic conditions. In addition, the multiclass model can capture the characteristics of mixed traffic flow such as the difference in value of time between HDVs and CAVs and the asymmetry in their driving interactions, thereby enhancing behavioral realism in the modeling. The study develops a new solution algorithm labeled RSRS-MSRA, in which a route-swapping based strategy is embedded with a self-regulated step size choice technique, to solve the proposed model efficiently. Sensitivity analysis of the proposed model is performed to gain insights into the effects of perturbations on the mixed traffic equilibrium, which facilitates the estimation of equilibrium traffic flow and identification of critical elements under expected or unexpected events. The study results can assist transportation decision-makers to design effective planning and operational strategies to leverage the advantages of CAVs and manage traffic congestion under mixed traffic flows.

Distributionally robust optimization of an emergency medical service station location and sizing problem with joint chance constraints

Abstract: An effective Emergency Medical Service (EMS) system can provide medical relief supplies for common emergencies (fire, accident, etc.) or large-scale disasters (earthquake, tsunami, bioterrorism attack, explosion, etc.) and decrease morbidity and mortality dramatically. This paper proposes a distributionally robust model for optimizing the location, number of ambulances and demand assignment in an EMS system by minimizing the expected total cost. The model guarantees that the probability of satisfying the maximum concurrent demand in the whole system is larger than a predetermined reliability level by introducing joint chance constraints and characterizes the expected total cost by moment uncertainty based on a data-driven approach. The model is approximated as a parametric second-order conic representable program. Furthermore, a special case of the model is considered and converted into a standard second-order cone program, which can be efficiently solved with a proposed outer approximation algorithm. Extensive numerical experiments are conducted to illustrate the benefit of the proposed approach. Moreover, a dataset from a real application is also used to demonstrate the application of the data-driven approach.

Stochastic bus schedule coordination considering demand assignment and rerouting of passengers

Abstract: Schedule coordination is a proven strategy to improve the connectivity and service quality for bus networks, whereas current research mostly optimizes schedule design using the a priori knowledge of users’ routings and ignores the behavioural reactions to coordination status. This study proposes a novel stochastic bus schedule coordination design with demand assignment and passenger rerouting in case of transfer failure. To this end, we develop a bi-level programming model in which the schedule design (headways and slack times) and passenger route choice are determined simultaneously via two travel strategies: non-adaptive and adaptive routings. In the second strategy, transfer passengers would modify their paths in case of missed connection. In this way, the expected network flow distribution is dependent on both the transfer reliability and network structure. The upper-level problem is formulated as a mixed integer non-linear program with the objective of minimizing the total system cost, including both operation cost and user cost, while the lower-level problem is route choice (pre-trip and on-trip) model for timed-transfer service. A more generalized inter-ratio headways scenario is also taken into account. A heuristic algorithm and the method of successive averages are comprehensively applied for solving the bi-level model. Results show that when the rerouting behaviour is considered, more cost-effective schedule coordination scheme with less slack times can be achieved, and ignoring such effect would underestimate the efficacy of schedule coordination scheme.

Managing connected and automated vehicles at isolated intersections: From reservation- to optimization-based methods

Abstract: Reservation-based methods with simple policies such as first-come-first-service (FCFS) have been proposed in the literature to manage connected and automated vehicles (CAVs) at isolated intersections. However, a comprehensive analysis of intersection capacity and vehicle delay under FCFS-based control is missing, especially under high traffic demand. To address this problem, this study adopts queueing theory and analytically shows that such method is incapable of handling high demand with multiple conflicting traffic streams. Furthermore, an optimization model is proposed to optimally serve CAVs arriving at an intersection for delay minimization. This study then compares the performance of the proposed optimization-based control with reservation-based control as well as conventional vehicle-actuated control at different demand levels. Simulation results show that the proposed optimization-based control performs best and it has noticeable advantages over the other two control methods. The advantages of reservation-based control are insignificant compared with vehicle-actuated control under high demand.

A branch-and-price algorithm for the heterogeneous fleet green vehicle routing problem with time windows

Abstract: Heterogeneous fleet vehicles can be used to reduce carbon emissions. We propose an improved branch-and-price (BAP) algorithm to precisely solve the heterogeneous fleet green vehicle routing problem with time windows (HFGVRPTW). In the improved BAP, to speed up the solution for the pricing problem, we develop a multi-vehicle approximate dynamic programming (MVADP) algorithm that is based on the labeling algorithm. The MVADP algorithm reduces labels by integrating the calculation of pricing problems for all vehicle types. In addition, to rapidly obtain a tighter upper bound, we propose an integer branch method. For each branch, we solve the master problem with the integer constraint by the CPLEX solver using the columns produced by column generation. We retain the smaller of the obtained integer solution and the current upper bound, and the branches are thus reduced significantly. Extensive computational experiments were performed on the Solomon benchmark instances. The results show that the branches and computational time were reduced significantly by the improved BAP algorithm.

A Markov decision process approach to vacant taxi routing with e-hailing

Abstract: The optimal routing of a vacant taxi is formulated as a Markov Decision Process (MDP) problem to account for long-term profit over the full working period. The state is defined by the node at which a vacant taxi is located, and action is the link to take out of the node. State transition probabilities depend on passenger matching probabilities and passenger destination probabilities. The probability that a vacant taxi is matched with a passenger during the traversal of a link is calculated based on temporal Poisson arrivals of passengers and spatial Poisson distributions of competing vacant taxis. Passenger destination probabilities are calculated directly using observed fractions of passengers going to destinations from a given origin. The MDP problem is solved by value iteration resulting in an optimal routing policy, and the computational efficiency is improved by utilizing parallelized matrix operations. The proposed model and an efficient implementation of the value iteration algorithm are tested in a case study with parameters derived from GPS trajectories of over 12,000 taxis in Shanghai, China for a study period of 5:30 - 11:30 am on a typical weekday. The optimal routing policy is compared with three heuristics based on simulated trajectories. Results show that the optimal routing policy improves average unit profit by 23.0% and 8.4% over the random walk and local hotspot heuristic respectively; and improves occupancy rate by 23.8% and 8.3% respectively. The improvement is larger during higher demand periods.

Two-phase optimal solutions for ship speed and trim optimization over a voyage using voyage report data

Abstract: In the daily operations of a shipping line, minimization of a ship's bunker fuel consumption over a voyage comprising a series of waypoints by adjusting its sailing speeds and trim settings plays a critical role in ship voyage management. To quantify the synergetic influence of sailing speed, displacement, trim, and weather and sea conditions on ship fuel efficiency, we first develop a tailored method to build two artificial neural network models using ship voyage report data. We proceed to address the ship sailing speed and trim optimization problem by putting forward three viable countermeasures within an effective two-phase optimal solution framework: sailing speeds of the ship are optimized in an on-shore planning phase, whereas trim optimization is conducted dynamically by the captain in real time when she/he observes the actual weather and sea conditions at sea. In the on-shore speed optimization problem, simultaneous optimization of sailing speeds and trim settings is beneficial in suggesting more informed sailing speeds because both factors influence a ship's fuel efficiency. In the countermeasure 3 proposed by this study, we address speed and trim optimization simultaneously by proposing a two-step global optimization algorithm that combines dynamic programming and a state-of-the-art simulation-based optimization technique. Numerical experiments with two 9000-TEU (twenty-foot equivalent unit) containerships show that (a) the proposed countermeasure 1 saves 4.96% and 5.83% of bunker fuel for the two ships, respectively, compared to the real situation; (b) the proposed countermeasure 2 increases the bunker fuel savings to 7.63% and 7.57%, respectively; and (c) the bunker fuel savings with Countermeasure 3 attain 8.25% on average. These remarkable bunker fuel savings can also translate into significant mitigation of CO2 emissions.

Model and analysis of labor supply for ride-sharing platforms in the presence of sample self-selection and endogeneity

Abstract: With the popularization of ride-sharing services, drivers working as freelancers on ride-sharing platforms can design their schedules flexibly. They make daily decisions regarding whether to participate in work, and if so, how many hours to work. Factors such as hourly income rate affect both the participation decision and working-hour decision, and evaluation of the impacts of hourly income rate on labor supply becomes important. In this paper, we propose an econometric framework with closed-form measures to estimate both the participation elasticity (i.e., extensive margin elasticity) and working-hour elasticity (i.e., intensive margin elasticity) of labor supply. We model the sample self-selection bias of labor force participation and endogeneity of income rate and show that failure to control for sample self-selection and endogeneity leads to biased estimates. Taking advantage of a natural experiment with exogenous shocks on a ride-sharing platform, we identify the driver incentive called “income multiplier” as exogenous shock and an instrumental variable. We empirically analyze the impacts of hourly income rates on labor supply along both extensive and intensive margins. We find that both the participation elasticity and working-hour elasticity of labor supply are positive and significant in the dataset of this ride-sharing platform. Interestingly, in the presence of driver heterogeneity, we also find that in general participation elasticity decreases along both the extensive and intensive margins, and working-hour elasticity decreases along the intensive margin.

Railway timetable rescheduling with flexible stopping and flexible short-turning during disruptions

Abstract: Railway operations are vulnerable to unexpected disruptions that should be handled in an efficient and passenger-friendly way. To this end, we propose a timetable rescheduling model where flexible stopping (i.e. skipping stops and adding stops) and flexible short-turning (i.e. full choice of short-turn stations) are innovatively integrated with three other dispatching measures: retiming, reordering, and cancelling. The Mixed Integer Linear Programming model also ensures that each train serving a station is ensured with a platform track. To consider the rescheduling impact on passengers, the weight of each decision is estimated individually according to the time-dependent passenger demand. The objective is minimizing passenger delays. A case study is carried out for hundreds of disruption scenarios on a subnetwork of the Dutch railways. It is found that (1) applying a mix of flexible stopping and flexible short-turning results in less passenger delays; (2) shortening the recovery duration mitigates the post-disruption consequence by less delay propagation but is at the expense of more cancelled train services during the disruption; and (3) the optimal rescheduling solution is sensitive to the disruption duration, but some steady behaviour is observed when the disruption duration increases by the timetable cycle time.

Recoverable robustness in weekly berth and quay crane planning

Abstract: The performance of a container terminal heavily relies on how efficiently the quayside resources, which are mainly berth and quay cranes, are used. The quayside related planning problems face uncertainty in various parameters, and this makes the efficient planning of these operations even more complicated. This study aims at developing a recoverable robust optimization approach for the weekly berth and quay crane planning problem. In order to build systematic recoverable robustness, a proactive baseline schedule with reactive recovery costs has been suggested. The uncertainty of vessel arrivals and the fluctuation in the container handling rate of quay cranes are considered. The baseline schedule includes berthing positions, times and quay crane assignments for all vessels along with vessel-specific buffer times and buffer quay cranes. The problem also introduces recovery plans for each scenario. The objective is to minimize the cost of baseline schedule, the recovery costs from the baseline schedule and the cost of scenario solutions for different realizations of uncertain parameters. A mathematical model and an adaptive large neighborhood based heuristic framework are presented to solve the novel problem. Computational results point out the strength of the solution methods and practical relevance for container terminals.

Regulating TNCs: Should Uber and Lyft set their own rules?

Abstract: We evaluate the impact of three proposed regulations of transportation network companies (TNCs) like Uber, Lyft and Didi: (1) A minimum wage for drivers, (2) a cap on the number of drivers or vehicles, and (3) a per-trip congestion tax. The impact is assessed using a queuing theoretic equilibrium model which incorporates the stochastic dynamics of the app-based ride-hailing matching platform, the ride prices and driver wages established by the platform, and the incentives of passengers and drivers. We show that a floor placed under driver earnings can push the ride-hailing platform to hire more drivers and offer more rides, at the same time that passengers enjoy faster rides and lower total cost, while platform rents are reduced. Contrary to standard competitive labor market theory, enforcing a minimum wage for drivers benefits both drivers and passengers, and promotes the efficiency of the entire system. This surprising outcome holds for almost all model parameters, and it occurs because the wage floor curbs TNC labor market power. In contrast to a wage floor, imposing a cap on the number of vehicles hurts drivers, because the platform reaps all the benefits of limiting supply. The congestion tax has the expected impact: fares increase, wages and platform revenue decrease. We also construct variants of the model to briefly discuss platform subsidy, platform competition, and autonomous vehicles.

Integrating Lagrangian and Eulerian observations for passenger flow state estimation in an urban rail transit network: A space-time-state hyper network-based assignment approach

Abstract: In this study, we focus on one of practically important research problems of integrating Lagrangian and Eulerian observations for passenger flow state estimation in an urban rail transit network. The task is accomplished by using a triple of flow, density, and speed to construct a discretized passenger flow state, further constructing a space-time-state (STS) hyper network so that we can utilize a better defined three-dimensional solution space to integrate structurally heterogeneous data sources. The monitoring data include passenger transaction records and identification space-time samples observed over a possible range of a few hours from origins to destinations (Lagrangian observations), and time-dependent passenger counts collected at some key bottleneck locations (Eulerian observations). To describe the complex urban-rail passenger flow evolution, passenger traveling and fixed sensor state transition processes can be unified within a STS path representation. To estimate the consistent system internal states between two different types of observations, we formulate a hyper network-based flow assignment model in a generalized least squares estimation framework. For applications in large-scale transportation networks, we decompose the proposed model into three easy-to-solve sub-problems. The proposed model is applied to a real-world case based on the Beijing subway network with complete smart card data for each passenger at his/her origin and destination and time-dependent passenger counts in several key transfer corridors, while the specific space-time trajectories of all passengers and high-resolution time-dependent congestion levels at platforms, in trains, and in transfer corridors are estimated. This proposed passenger flow state inference method can provide a rich set of state inferences for advanced transit planning and management applications, for instance, passenger flow control, adaptive travel demand management, and real-time train scheduling.

Microscopic optimization model and algorithm for integrating train timetabling and track maintenance task scheduling

Abstract: This paper addresses the problem of improving the integration between passenger timetabling and track maintenance scheduling. We propose a microscopic optimization model and an iterative algorithm for solving this problem efficiently. Block sections are considered as the basic microscopic elements for train movements in a railway network. A mixed-integer linear programming formulation is proposed for the integrated optimization problem in which train timing, sequencing and routing are the timetabling variables, while timing and sequencing of maintenance tasks are the track maintenance variables. The objective function is to minimize the total train travel time and the maintenance tardiness cost. The constraints proposed in this work address the practical specifications of the INFORMS RAS 2016 Problem Solving Competition (2016 PSC). In this context, the main decision variables are the entrance and exit times of the trains on each block section plus the start and end times of each maintenance task. Since the integrated optimization problem is strongly NP-hard, an iterative algorithm is proposed to compute near-optimal solutions in a short computation time. The algorithm is based on a decomposition of the overall problem into sub-problems related to train scheduling and/or routing with or without track maintenance task scheduling. The connecting information between the two sub-problems concerns the selected train routes plus the start and end times of the maintenance tasks. Computational experiments are performed on a set of realistic railway instances, which were introduced during the 2016 PSC. The iterative algorithm outperforms a standard MILP solver and the first-place team of this competition in terms of both solution quality and time to deliver the new best-known solutions. The scalability of the iterative algorithm is investigated when increasing the number of trains and track maintenance tasks.

Vehicle assignment and relays for one-way electric car-sharing systems

Abstract: One-way electric car-sharing systems are expected to be an integral part of future transportation systems, playing an important role in reducing traffic congestion and carbon emissions. Owing to limited battery capacities and the lengthy charging process, an electric car-sharing system may not achieve the high vehicle utilization that a non-electric car-sharing system can achieve. We investigate two approaches to vehicle management, in order to circumvent battery constraints and to improve vehicle utilization rates in one-way electric car-sharing systems. In the first approach, we optimize vehicle assignment decisions, and in the second approach, we further allow vehicle relays, enabling users to complete longer trips by sequentially taking two vehicles. We propose a novel space–time–battery network flow model to determine optimal assignment and relay decisions. With an extra dimension for tracking each vehicle’s battery level, the proposed model is a single-commodity network flow model that is computationally amiable. To meet the requirement of quick responses, we propose an efficient algorithm that exploits an innovative diving heuristic to solve the resulting integer program. Numerical results show that when vehicle assignment and relays are optimized in an electric car-sharing system, we may achieve a comparable vehicle utilization rate as in a non-electric car-sharing system. In particular, optimizing vehicle assignment is essential when most trips are short trips, and vehicle relays are critical when users demand more long trips.

Integrated optimization of location, inventory and routing in supply chain network design

Abstract: We study the integrated optimization of location, inventory and routing in supply chain network design (SCDN) problem. The SCND decision has significant impact on the performance of the supply chain. The major obstacles of the integrated optimization are the high non-linearity and the huge amount of variables and constraints caused by routing. In this paper, we introduce real-world constraints into the integrated model to characterize the model much more precisely, which simultaneously reduce the dimension of the problem. We propose an exact algorithm based on the Generalized Benders Decomposition (GBD) method to solve the model. We can obtain the Benders Cuts (BC) explicitly. We verify the effectiveness of the model and method by a real-life case: A passenger car supply chain network design case.

Robust local and string stability for a decentralized car following control strategy for connected automated vehicles

Abstract: This paper presents a robust car-following control strategy under uncertainty for connected and automated vehicles (CAVs). The proposed control is designed as a decentralized linear feedback and feedforward controller with a focus on robust local and string stability under (i) time-varying uncertain vehicle dynamics and (ii) time-varying uncertain communication delay. The former uncertainty is incorporated into the general longitudinal vehicle dynamics (GLVD) equation that regulates the difference between the desired acceleration (prescribed by the control model) and the actual acceleration by compensating for nonlinear vehicle dynamics (e.g., due to aerodynamic drag force). The latter uncertainty is incorporated into acceleration information received from the vehicle immediately ahead. As a primary contribution, this study derives and proves (i) a sufficient and necessary condition for local stability and (ii) sufficient conditions for robust string stability in the frequency domain using the Laplacian transformation. Simulation experiments verify the correctness of the mathematical proofs and demonstrate that the proposed control is effective for ensuring stability against uncertainties.

The Electric Autonomous Dial-a-Ride Problem

Abstract: In the Dial-a-Ride-Problem (DARP) a fleet of vehicles provides shared-ride services to users specifying their origin, destination, and preferred arrival time. Typically, the problem consists of finding minimum cost routes, satisfying operational constraints such as time-windows, origin-destination precedences, user maximum ride-times, and vehicle maximum route-durations. This paper presents a problem variant for the DARP which considers the use of electric autonomous vehicles (e-ADARP). The problem covers battery management, detours to charging stations, recharge times, and selection of destination depots, along with classic DARP features. The goal of the problem is to minimize a weighted objective function consisting of the total travel time of all vehicles and excess ride-time of the users. We formulate the problem as a 3-index and a 2-index mixed-integer-linear program and devise a branch-and-cut algorithm with new valid inequalities derived from e-ADARP properties. Computational experiments are performed on adapted benchmark instances from DARP literature and on instances based on real data from Uber Technologies Inc. Instances with up to 5 vehicles and 40 requests are solved to optimality.

Modelling car-following behaviour of connected vehicles with a focus on driver compliance

Abstract: This paper incorporates the driver compliance behaviour into a connected vehicle driving strategy (CVDS) that can be integrated with traditional car-following (CF) models to better describe the connected vehicle CF behaviour. Driver compliance, a key human factor for the success of connected vehicles technology, is modelled using a celebrated theory of decision making under risk – the Prospect theory (PT). The reformulated value and weighting functions of PT are consistent with the driver compliance behaviour and also preserve the integral elements of PT. Furthermore, the connected vehicle trajectory data collected from a carefully designed advanced driving simulator experiment are utilised to calibrate CVDS integrated with Intelligent Driver Model (IDM), i.e., CVDS-IDM. The calibration results reveal that drivers in the connected environment drive safely and efficiently. Moreover, the CVDS-IDM can successfully model and predict the CF dynamics of connected vehicles and is more behaviourally and numerically sound than a traditional CF model.

A general model of demand-responsive transportation services: From taxi to ridesharing to dial-a-ride

Abstract: The paper presents a general analytic framework to model transit systems that provide door-to-door service. The model includes as special cases non-shared taxi and demand responsive transportation ( DRT ). In the latter we include both, paratransit services such as dial-a-ride ( DAR ), and the form of ridesharing ( shared taxi ) currently being used by crowd-sourced taxi companies like Lyft and Uber. The framework yields somewhat optimistic results because, among other things, it is deterministic and does not track vehicles across space. By virtue of its simplicity, however, the framework yields approximate closed form formulas for many cases of interest.

Solving cyclic train timetabling problem through model reformulation: Extended time-space network construct and Alternating Direction Method of Multipliers methods

Abstract: The cyclic train timetabling problem aims to synchronize limited operational resources toward a master periodic schedule of transport services. By introducing an extended time-space network construct, this paper proposes a new type of integer programming model reformulation for the cyclic train timetabling problem on a double-track railway corridor at the macroscopic level. This reformulation method also holds the promises to be applied in a broader set of routing and scheduling problems with periodic activity requirements. We also hope that this space-time network extension technique, as a special version of variable splitting methods in the dual decomposition literature, could potentially bridge the modeling gaps between cyclic and non-cyclic timetables. Specifically, the existing mathematical programming model for the periodic event scheduling problem (PESP) is transformed into a multi-commodity network flow model with two coupled schedule networks and side track capacity constraints. In addition, two dual decomposition methods including Lagrangian relaxation and Alternating Direction Method of Multipliers (ADMM), are adopted to dualize the side track capacity constraints. For each train-specific sub-problem in an iterative primal and dual optimization framework, we develop an enhanced version of forward dynamic programming to find the time-dependent least cost master schedule across the time-space network over multiple periods. ADMM-motivated heuristic methods with adjusted penalty parameters are also developed to obtain good upper bound solutions. Based on real-world instances from the Beijing-Shanghai high-speed railway corridor, we compare the numerical performance between the proposed reformulation and the PESP model that involves the standard optimization solver.

Modelling and managing the integrated morning-evening commuting and parking patterns under the fully autonomous vehicle environment

Abstract: This study is the first in the literature to analytically investigate the traffic dynamics of the integrated morning and evening commutes when daily trips are completed with autonomous vehicles (AVs). Given the parking locations of AVs resulting from the morning commute, firstly we analyse the evening commuting pattern, at which no traveller can reduce the individual travel cost given other AVs’ times of departures from the parking spaces. The equilibrium traffic pattern at the evening commute is then integrated with the morning commute, where equilibrium choices of departure time from home and parking location are derived and analysed. We then study the integrated morning-evening commuting pattern at the system optimum and develop the road tolling scheme to achieve the system optimum. Furthermore, this study analyses the optimal AV parking supply strategy to minimise the total system cost, which is comprised of the total social parking cost and the total daily travel cost under either user equilibrium or system optimum traffic pattern. We also illustrate the modelling insights through numerical studies regarding relationship among traffic efficiency, tolling schemes and AV parking supply plans. This study highlights the differences in daily commuting and parking patterns between the AV situation and the non-AV situation, and sheds light on how traffic and parking should be managed or planned in the future.

Development of a non-parametric classifier: Effective identification, algorithm, and applications in port state control for maritime transportation

Abstract: Maritime transportation plays a pivotal role in the economy and globalization, while it poses threats and risks to the maritime environment. In order to maintain maritime safety, one of the most important mitigation solutions is the Port State Control (PSC) inspection. In this paper, a data-driven Bayesian network classifier named Tree Augmented Naive Bayes (TAN) classifier is developed to identify high-risk foreign vessels coming to the PSC inspection authorities. By using data on 250 PSC inspection records from Hong Kong port in 2017, we construct the structure and quantitative parts of the TAN classifier. Then the proposed classifier is validated by another 50 PSC inspection records from the same port. The results show that, compared with the Ship Risk Profile selection scheme that is currently implemented in practice, the TAN classifier can discover 130% more deficiencies on average. The proposed classifier can help the PSC authorities to better identify substandard ships as well as to allocate inspection resources.

Robust location-allocation network design for earthquake preparedness

Abstract: We develop robust models for earthquake preparedness by optimizing the number, location, and capacity of distribution centers (DCs). The goal is to minimize the total social costs, which include setup and initial supplies, as well as the deprivation costs associated with delayed access to supplies. The models incorporate various earthquake magnitude-specific uncertainties, such as facility damage, casualty by severity, and travel time. Examining the concept of social costs in light of an emerging concern in humanitarian logistics - the robustness of relief networks, we model two types of robustness: parameter uncertainty within a scenario and relative regret across scenarios. This unique approach reveals (1) the magnitude of social costs in the aftermath of an earthquake; (2) the hidden risks associated with inaccurate modeling of deprivation costs; and (3) the impact of budgetary constraints. We demonstrate the applicability of our approach via a case study featuring the Northridge region in California, which experienced two of the strongest earthquakes recorded in North America in 1971 and 1994.