Large-scale Transportation Systems Research Articles

A new multi-input multi-output control design method for a large-scale web transport system

The paper introduces an innovative multi-input multi-output (MIMO) control design approach tailored for the complex dynamics of large-scale web transport systems in printed electronics manufacturing technology. First, a novel hybrid control design is proposed to a MIMO strict-feedback nonlinear dynamic system, in which a modified tracking error is introduced in term of a sliding surface for design purpose. Then, a combined use of Lyapunov-based control design and sliding mode control is employed to formulate a novel control law, called Backstepping Sliding Mode Control (BSMC) and a BSMC based control structure is provided. Second, application of this control strategy is implemented to control web tension and velocity of a three-span web transport system. In this study, a series of assumptions are provided to transform an original system dynamics of web transport system into a standard strict-feedback form. Then, the proposed theory is applied for the achieved system, a BSMC for the system is synthesized for simulation study. Finally, numerical simulations are conducted under different operating conditions to validate the effectiveness, reliability, and robustness of the proposed control method.

Estimating Markov Chain Mixing Times: Convergence Rate Towards Equilibrium of a Stochastic Process Traffic Assignment Model

Network equilibrium models have been extensively used for decades. The rationale for using equilibrium as a predictor is essentially that (i) a unique and globally stable equilibrium point is guaranteed to exist and (ii) the transient period over which a system adapts to a change is sufficiently short in time that it can be neglected. However, we find transport problems without a unique and stable equilibrium in the literature. Even if it exists, it is not certain how long it takes for the system to reach an equilibrium point after an external shock onto the transport system, such as infrastructure improvement and damage by a disaster. The day-to-day adjustment process must be analysed to answer these questions. Among several models, the Markov chain approach has been claimed to be the most general and flexible. It is also advantageous as a unique stationary distribution is guaranteed in mild conditions, even when a unique and stable equilibrium does not exist. In the present paper, we first aim to develop a methodology for estimating the Markov chain mixing time (MCMT), a worst-case assessment of the convergence time of a Markov chain to its stationary distribution. The main tools are coupling and aggregation, which enable us to analyse MCMTs in large-scale transport systems. Our second aim is to conduct a preliminary examination of the relationships between MCMTs and critical properties of the system, such as travellers’ sensitivity to differences in travel cost and the frequency of travellers’ revisions of their choices. Through analytical and numerical analyses, we found key relationships in a few transport problems, including those without a unique and stable equilibrium. We also showed that the proposed method, combined with coupling and aggregation, can be applied to larger transport models. History: This paper has been accepted for the Transportation Science Special Issue on the 25th International Symposium on Transportation and Traffic Theory. Funding: This study was financially supported by the Japan Society for the Promotion of Science [Grant-in-Aid 20H00265]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/trsc.2024.0523 .

Imbalanced data augmentation for pipeline fault diagnosis: A multi-generator switching adversarial network

Pipeline fault diagnosis is of practical necessity for the stable operation of large-scale oil and gas transportation systems. Despite the remarkable achievements of existing intelligent fault diagnosis methods, these studies have encountered noticeable performance degradation when directly applied in real industrial scenarios due to the absence of sufficient and balanced datasets. To this end, this paper proposes a novel multi-generator switching adversarial network (MGSAN) to augment the pipeline data under different health states, especially faulty ones. First, a multi-mode generator is designed to produce various candidate solutions, where a parameter-sharing scheme is employed to ensure consistency among modes at the parameter level. After that, a mode switching strategy is presented to maintain a proper tradeoff between the generator and the discriminator, which is capable of changing the state of the generator from one mode to another according to the performance (i.e., quality and diversity) of the candidate solutions. Finally, extensive experimental results on visualization and quantitative evaluation show that the MGSAN algorithm has more stable behavior and stronger exploration capabilities than well-known GAN variants. Additionally, the proposed MGSAN algorithm is successfully exploited to improve the accuracy of pipeline fault diagnosis.

Optimal policy for scheduling automated guided vehicles in large-scale intelligent transportation systems

Automated electric mobility technologies have been increasingly applied to large-scale intelligent transportation systems (ITSs) to enhance productivity and efficiency. Advanced technologies have reshaped the traditional transportation systems and posed numerous challenges to the deployment and management of new ITSs. A major challenge in the real-life implementation of ITSs is how to manage a large number of automated objects in a cooperative manner. In this paper, we investigate the scheduling and routing problem of automated guided vehicles (AGVs) in a complicated ITS. Cost and efficiency are identified as the two crucial performance indicators of such a novel ITS. An easy-to-implement practical decision policy and a tailored particle swarm based solution method are designed for problem solving. In addition to the theoretical contributions, this paper also conducts a case study to validate the effectiveness and applicability of the proposed methodology, thus contributing to the planning and management of large-scale transportation systems by modeling, optimizing, and validating a new ITS deployed with AGVs.

Identifying the cargo types of road freight with semi-supervised trajectory semantic enhancement

Identifying road freight cargo types is crucial for regional economic interaction and transportation optimization. Existing methods primarily rely on manual labeling and the rule, neither of which can achieve automated semantic enhancement of large-scale road freight trajectories. Consequently, this study proposes a semi-supervised trajectory semantic enhancement method for identifying cargo types based on trajectory feature extraction and point-of-interest (POI) association. The raw trajectories are segmented and enriched with the closest POIs. The sample labeling method with POI semantic enhancement is then proposed using company registration information. Finally, the spatiotemporal and sequential features of labeled freight trips are extracted to build a self-training semi-supervised model for identifying the cargo type of road freight. Experimental studies on real trajectory data demonstrate superior accuracy and robustness compared to existing methods, with accuracy and F1 values reaching 81.4 and 0.77%, respectively. The proposed sample labeling method improves representativeness and universality, increasing accuracy by 7.8–14.4% and F1 value by 8.5–34.5% compared to the rule-based method. The semi-supervised model improves accuracy by 8.9% and F1 value by 29.1% compared to the supervised model when only 10.0% of samples were labeled. This method enables automatic and full-sample cargo type identification in real-world large-scale transportation systems.

Optimal allocation of fast charging stations for large-scale transportation systems

The modern quest for sustainable cities increasingly relies on using distributed energy resources (DERs), which requires new planning practices. This paper proposes an optimisation strategy to solve the fast charging station (FCS) allocation of electric vehicles (EVs). A mixed-integer programming (MIP) model minimises investment and operation costs, considering the building of FCSs with photovoltaic (PV) systems over carports and battery energy storage systems (BESSs) as planning alternatives. The embedded set covering problem has special aspects that allow the development of a novel approach to evaluate candidate sites to accommodate FCSs. A preprocessing strategy is developed to fine-tune the entire solution space. A multiobjective approach is used to obtain an optimal compromise solution for the MIP model when it is required to serve the maximum number of EV owners at the lowest possible cost. The combined strategies reduce the computational burden, allowing full-scale studies of EV charging system planning. The results of studies using a real-world Brazilian case certify the benefits of the proposed strategy in the FCS allocation problem and in optimising the operation when considering renewable alternatives.

Multi-Level Progressive Learning for Unsupervised Vehicle Re-Identification

Vehicle re-identification (ReID) technology has played a more and more important role in Intelligent Transport System (ITS), which aims at searching the same query vehicle identity from a large amount of gallery datasets under different non-overlapping camera views. Current related researches mainly focus on discriminative feature mining of vehicle images and train the model in a fully supervised manner which highly relies on the manual annotations of training data. However, it is labor-consuming and impractical to generate the annotation for each sample image in real-word applications especially for those large-scale transport systems with tons of surveillance data. To this point, we propose in this paper a multi-level progressive learning (MLPL) method for unsupervised vehicle ReID, which gives a good performance by only utilizing the unlabeled target domain images. We firstly introduce a multi-branch architecture to explore the vehicle representations in different level, which consists of one branch for global feature and two branches for local feature learning. A density-based clustering method is employed to generate pseudo labels. Combining with the unique model, we propose a novel re-clustering method to better mine the labels with high reliability. Then a dynamic progressive contrast learning (DPCL) strategy is carefully designed to train the network based on these clustered labels. DPCL could dynamically adjust the training process to maximally strengthen the multi-level feature learning. Moreover, we further propose a self-adaptive loss balance method to automatically compute the weights of different losses during each training iteration. Comprehensive experiments are conducted on several mainstream evaluation datasets, including VeRi776, VehicleID and CityFlowV2-ReID. Compared to other existed unsupervised methods, our approach achieves the new state-of-the-art performance.

Bus frequency optimization in a large-scale multi-modal transportation system: integrating 3D-MFD and dynamic traffic assignment

Bus frequency optimization in a large-scale multi-modal transportation system: integrating 3D-MFD and dynamic traffic assignment

A freight asset choice model for agent-based simulation models

Agent-based models (ABM) for transportation operations have been largely focused on passenger trips, while recent developments in the field began the incorporation of freight-related operations. However, these models rarely incorporate freight-related strategic asset decisions: fleet and distribution center (DC) ownership. These attributes are important for modeling freight transportation behavior in ABMs. This research develops behavioral models that jointly predict fleet ownership and distribution center control for freight-related firms in an ABM framework. A seemingly unrelated Tobit regression is estimated using large-scale data from more than 11 million establishments. The model is estimated using a Bayesian approach that allows for the quantification of the coefficients’ variability. Model results indicate that firms with higher revenue have an increased propensity to own fleet and (or) DC and prefer larger fleets and more DC space. Furthermore, Transportation firms generally have more heavy-duty trucks and much fewer medium-duty trucks, while firms in all other sectors strongly prefer medium-duty fleets. Food Services and other Retail firms have the greatest preference for owning or leasing their own DCs, followed by Manufacturing, Wholesale, and Transportation firms. A case study is developed for the city of Chicago using a high-performance ABM framework designed for simulating large-scale transportation systems. Transportation modelers and policymakers can use findings and methods from this research to study freight operations in large cities.

Hybrid Discrete Differential Evolution and Deep Q-Network for Multimission Selective Maintenance

The multimission selective maintenance problem (MSMP) for repairable systems has received increasing attention in recent years. The problem amounts to selecting a subset of feasible maintenance actions, in view of the resource limitations. For considering the realistic case of the imperfect maintenance, this article introduces a hybrid imperfect maintenance model, which is more realistic to evaluate the system reliability. The challenge of solving such kind of problems lies not only in the reliability estimation, but also in the solution method of the maintenance selection. Such decision-making problem can be effectively formulated using the Markov decision process, but it is difficult to apply current methods for solving the engineering systems with large action decision spaces. In order to solve this issue, this work puts forth a novel hybrid algorithm for the MSMP in a large multicomponent system. In the proposed method, a discrete differential evolution algorithm is developed for searching the optimal maintenance action in large-scale discrete action spaces and the deep Q-network method is utilized to approximate the effectiveness of maintenance actions and facilitate the agent training. The experiments, based on a large-scale coal transportation system, verify the effectiveness of the proposed method compared with LSDQN and differential evolution.

Large-scale long-distance land-based hydrogen transportation systems: A comparative techno-economic and greenhouse gas emission assessment

Interest in hydrogen as an energy carrier is growing as countries look to reduce greenhouse gas (GHG) emissions in hard-to-abate sectors. Previous works have focused on hydrogen production, well-to-wheel analysis of fuel cell vehicles, and vehicle refuelling costs and emissions. These studies use high-level estimates for the hydrogen transportation systems that lack sufficient granularity for techno-economic and GHG emissions analysis. In this work, we assess and compare the unit costs and emission footprints (direct and indirect) of 32 systems for hydrogen transportation. Process-based models were used to examine the transportation of pure hydrogen (hydrogen pipeline and truck transport of gaseous and liquified hydrogen), hydrogen-natural gas blends (pipeline), ammonia (pipeline), and liquid organic hydrogen carriers (pipeline and rail). We used sensitivity and uncertainty analyses to determine the parameters impacting the cost and emission estimates. At 1000 km, the pure hydrogen pipelines have a levelized cost of $0.66/kg H2 and a GHG footprint of 595 gCO2eq/kg H2. At 1000 km, ammonia, liquid organic hydrogen carrier, and truck transport scenarios are more than twice as expensive as pure hydrogen pipeline and hythane, and more than 1.5 times as expensive at 3000 km. The GHG emission footprints of pure hydrogen pipeline transport and ammonia transport are comparable, whereas all other transport systems are more than twice as high. These results may be informative for government agencies developing policies around clean hydrogen internationally.

Modeling and Optimizing Large-Scale Production-Level Transportation Systems

Modeling and Optimizing Large-Scale Production-Level Transportation Systems

Bus Frequency Optimization in a Large-Scale Multi-Modal Transportation System: Integrating 3d-Mfd and Dynamic Traffic Assignment

Bus Frequency Optimization in a Large-Scale Multi-Modal Transportation System: Integrating 3d-Mfd and Dynamic Traffic Assignment

Dynamic Ride-Matching for Large-Scale Transportation Systems

Efficient dynamic ride-matching (DRM) in large-scale transportation systems is a key driver in transport simulations to yield answers to challenging problems. Although the DRM problem is simple to solve, it quickly becomes a computationally challenging problem in large-scale transportation system simulations. Therefore, this study thoroughly examines the DRM problem dynamics and proposes an optimization-based solution framework to solve the problem efficiently. To benefit from parallel computing and reduce computational times, the problem’s network is divided into clusters utilizing a commonly used unsupervised machine learning algorithm along with a linear programming model. Then, these sub-problems are solved using another linear program to finalize the ride-matching. At the clustering level, the framework allows users adjusting cluster sizes to balance the trade-off between the computational time savings and the solution quality deviation. A case study in the Chicago Metropolitan Area, U.S., illustrates that the framework can reduce the average computational time by 58% at the cost of increasing the average pick up time by 26% compared with a system optimum, that is, non-clustered, approach. Another case study in a relatively small city, Bloomington, Illinois, U.S., shows that the framework provides quite similar results to the system-optimum approach in approximately 62% less computational time.

Reconnecting Eurasia: a new logistics state, the China–Europe freight train, and the resurging ancient city of Xi’an

ABSTRACT Large-scale transport systems project expansive geographical reach via far-reaching connectivity and spillovers. This phenomenon, however, is understudied for its impact on economic and spatial relations across geographic scales and economic domains and the mechanism carrying and transmitting that impact. Despite its short existence, the China–Europe Freight Train (CEFT) has already created a long geographical reach and major impact on the transport landscape spanning China, Central Asia, and Europe. This paper argues that a new logistics state in China at the local level is driving and sustaining the CEFT from below relative to the national government and market forces. Using the ancient city of Xi’an as a characteristic embodiment of a logistics state, this paper demonstrates how the logistics state-driven CEFT has multiplied routes and redirected trade flows between China and Europe, reorganized inter-city and cross-border production and supply chains from China to Europe, stimulated a new geography of globally oriented and nationally rebalanced local consumption in China, and fueled major new development in an ancient and economically lagging city. The paper concludes on critical complications and implications from the Chinese local logistics state for the CEFT’s sustainability and future research.

Evolution towards optimal driving strategies for large‐scale autonomous vehicles

With rapidly developing autonomous vehicle (AV) technologies, the optimal driving strategy should consider multi-objective optimization problems of large-scale transportation systems, including safety and efficiency. Different driving strategies have different performance, and there is an interaction between vehicles with different strategies. Since the Nash equilibrium is hard to find for an n-player game, it is difficult to get an analytical solution to this multi-objective optimization problem. Therefor a coevolutionary algorithm is proposed to explore the interactions between populations with different strategies and investigate the cooperation and competition among vehicles. Combining the multi-objective optimization algorithm and the incentive mechanism of survival of the fittest reproduction law, the system structure reaches a stable equilibrium state and the optimal group driving strategy evolves. Simulation results, with 40,000 vehicles driving in Luxembourg SUMO Traffic Scenario, demonstrate that the rational–rational strategy performs best among six typical strategies. Meanwhile, the accident rate drops by 56%, while the overall average speed increases by 30%. The results of multi-vehicle and multi-objective coevolution are enlightening in designing optimal driving strategy with AVs.

Large-scale transportation and storage of wood pellets: Investigation of the change in physical properties

The change in physical properties of wood pellets, with a focus on particle size distributions due to pellet breakage and attrition, was studied in a large-scale (∼450ton/h) transportation system. Critical locations with a high probability of breakage through the whole transportation system were chosen and sampled to study the effect of transportation system design and operation on the mechanical properties of pellets. Bulk density, mechanical durability, moisture content, and particle size distribution of pellets were characterized for each sample. Analysis of variance showed that there were significant differences between the percentages of small particles (< 5.6mm) in the samples taken at different locations, especially at one with a vertical free fall of 7.8m. On average, this relatively long drop increased the proportion of particles < 5.6mm in the samples from 8.73% to 14.09%, and that of particles < 3.15mm from 4.82% to 9.01%. Moreover, the measurements showed a wide deviation in the mechanical durability values, between a minimum of 90.8% and a maximum of 98.7%, which were not correlated to the sampling points but related to pellet properties. It can be concluded that pellet transportation systems require more dedicated design strategies to prevent breakage and attrition.

Impact of High-Speed Rail on Road Traffic and Greenhouse Gas Emissions

Carbon emission reduction in the transportation sector is essential in the global mitigation effort and a large-scale public transport system has the potential to be an effective instrument. High speed rail (HSR) is one such example, yet it is unclear how much reduction in road traffic results from new rail routes. Using the difference-in-differences (DID) method, we show that new HSR routes in China leads to a 20.5 log-point reduction in the number of passenger vehicles and a 15.7 log-point reduction in freight vehicles running on parallel highways. These reductions were not seen on ordinary national roads. These effects translate into an annual reduction of 14.76 million tons of CO2 equivalent of greenhouse gas (GHG) emissions, or 1.75% of GHG emissions in China’s transport sector. This mitigation effect mainly comes from the substitution of highway goods transport with the conventional railway, instead of the direct replacement of highway passenger transport with HSR. The environmental benefit of HSR in China has not been fully realized because of the thermal-dominated electricity supply. Our further projections suggest that in greener electricity conditions, the HSR in China can significantly contribute more to the reduction in GHG emissions from the transport sector.

Recording Permanence and Ephemerality in the North Quarter of Brussels: Drawing at the Intersection of Time, Space, and People

Lying in the Senne River Valley, the North Quarter of Brussels is a physical record of spatial transformations unevenly distributed over time. Waves of developments and unfinished plans colonized its original landscape structure, erasing, writing, and re-writing it with large-scale metropolitan projects and transportation systems, around which an industrial and urban fabric developed. Accumulated expansions left an assemblage of incomplete infrastructures in which a multi-faceted and highly identifiable quarter lies punctuated by weakly defined morphological mismatches. At the center of this diverse and mutilated fabric, Maximilien Park stands as pars pro toto. From a combination of research methods that includes ethnographic fieldwork and interpretative mapping, three drawings are overlaid with the moving dimensions of space, time, and people, and assembled in a reinterpreted triptych to investigate the production of that public space. The first panel “Traces” overlaps lost urban logics and remaining traces on the urban tissue. The second panel “Cycles” traces the uneven deconstruction of the North Quarter during the last century, identifying scars of its past. The third panel “Resignifications” focuses on recent events in the area, examining how people have appropriated and transformed the park since 2015. With this triptych, the article aims to re-interpret the palimpsest of the North Quarter, represent the area’s transforming character, and unravel a spatial reading of the lived experiences of the place through time.

Integrating Supply and Demand Perspectives for a Large-Scale Simulation of Shared Autonomous Vehicles

Transportation Network Companies (TNCs) have been steadily increasing the share of total trips in metropolitan areas across the world. Micro-modeling TNC operation is essential for large-scale transportation systems simulation. In this study, an agent-based approach for analyzing supply and demand aspects of ride-sourcing operation is done using POLARIS, a high-performance simulation tool. On the demand side, a mode-choice model for the agent and a vehicle-ownership model that informs this choice are developed. On the supply side, TNC vehicle-assignment strategies, pick-up and drop-off operations, and vehicle repositioning are modeled with congestion feedback, an outcome of the mesoscopic traffic simulation. Two case studies of Bloomington and Chicago in Illinois are used to study the framework’s computational speed for large-scale operations and the effect of TNC fleets on a region’s congestion patterns. Simulation results show that a zone-based vehicle-assignment strategy scales better than relying on matching closest vehicles to requests. For large regions like Chicago, large fleets are seen to be detrimental to congestion, especially in a future in which more travelers will use TNCs. From an operational point of view, an efficient relocation strategy is critical for large regions with concentrated demand, but not regulating repositioning can worsen empty travel and, consequently, congestion. The TNC simulation framework developed in this study is of special interest to cities and regions, since it can be used to model both demand and supply aspects for large regions at scale, and in reasonably low computational time.