Real Transportation Network Research Articles

AlarmGPT: an intelligent alarm analyzer for optical networks using a generative pre-trained transformer

The proliferating development of optical networks has broadened the network scope and caused a corresponding rise in equipment deployment. This growth potentially results in a significant number of alarms in the case of equipment malfunctions or broken fiber. Managing these alarms efficiently and accurately has always been a critical concern within the research and industry community. The alarm processing workflow typically includes filtration, analysis, and diagnostic stages. In current optical networks, these procedures are often performed by experienced engineers, utilizing their expert knowledge and extensive experience. This method requires considerable human resources and time, as well as demanding proficiency prerequisites. To address this issue, we propose an intelligent alarm analysis assistant, “AlarmGPT,” for optical networks, utilizing a generative pre-trained transformer (GPT) and LangChain. The proposed AlarmGPT exhibits a high level of semantic comprehension and contextual awareness of alarm data, significantly enhancing the model’s ability of interpreting, classifying, and solving alarm events. Through verification of extensive alarm data collected from real optical transport networks (OTNs), the usability of AlarmGPT has been validated in the tasks of alarm knowledge Q&A, alarm compression, alarm priority analysis, and alarm diagnosis. This method has the potential to significantly reduce the labor and time required for alarm processing, while also lowering the experiential requisites incumbent upon network operators.

Symmetry breaking in optimal transport networks

Engineering multilayer networks that efficiently connect sets of points in space is a crucial task in all practical applications that concern the transport of people or the delivery of goods. Unfortunately, our current theoretical understanding of the shape of such optimal transport networks is quite limited. Not much is known about how the topology of the optimal network changes as a function of its size, the relative efficiency of its layers, and the cost of switching between layers. Here, we show that optimal networks undergo sharp transitions from symmetric to asymmetric shapes, indicating that it is sometimes better to avoid serving a whole area to save on switching costs. Also, we analyze the real transportation networks of the cities of Atlanta, Boston, and Toronto using our theoretical framework and find that they are farther away from their optimal shapes as traffic congestion increases.

A novel approach to traffic modelling based on road parameters, weather conditions and GPS data using feedforward neural networks

This article presents the development of a estimation model using an artificial neural network to estimate the traffic factor parameter, which reflects changes in travel time for a single transport connection. The modeling of traffic is highly challenging due to the complex and non-linear nature of road systems, the interactions between diverse road users, and the ripple effects of congestion. The research problem addressed is the need to consider a wide range of factors that contribute to congestion, as identified in the literature. To address this challenge, a universal Feedforward Neural Network (FNN) was designed, specifically the Multilayer Perceptron (MLP) 38–10-1 model. The input layer neurons of the MLP receive a vector of input values representing various factors such as road types, technical properties, time of day, days of the week, incidents, weather conditions, and population density. The model was trained using a dataset consisting of 41,945 records of fully completed input variables, extracted from a larger dataset of 300,000 travel time measurements collected from a real transport network mapped on a macro scale (Mazovia, Poland). During the research, the model achieved a satisfactory level of Mean Absolute Percentage Error (MAPE) for the test set (9.12%) and a Normalized Root Mean Squared Error (NRMSE) below 0.02, indicating good estimation performance. Furthermore, this work proposes specialized models with the structure of MLP 34–10-1 for individual types of roads, which also demonstrated satisfactory estimation abilities. These models have practical applications in traffic management and planning. Overall, this research addresses the research problem of modeling factors for predicting travel time changes using an artificial neural network approach. The developed models provide accurate estimations and offer potential applications in transportation systems.

ESTIMATION AND MAPPING OF THE SETTLEMENT FIELD POTENTIAL BASING ON REAL TRANSPORTATION CONNECTIONS BETWEEN SETTLEMENTS

Abstract. In the article, we describe an approach to computational estimation of the settlement field potential, and mapping its spatial distribution in the desktop geographic information system software. The settlement field potential is a quantitative geographical variable that reflects mutual impact of settlements according to the population amount in the settlements and distances between settlements (two or many). This geographical variable can be used as an extra metric in human geography alongside with population density metric. We propose a methodology for automation of the settlement field potential mapping, and propose to estimate the settlement field potential value using distances measured in real transportation network instead of traditionally used straight-line measurement of the distances. To test and verify proposed approach and methodology we produced the maps of settlement field potential for the study area that covers Russia-Kazakhstan transborder area.

Modeling elasticity, similarity, stochasticity, and congestion in a network equilibrium framework using a paired combinatorial weibit choice model

In the traffic assignment problem for predicting traffic flow patterns in a transportation network, it is important to account for route overlap and non-identical perception variance in route choice analysis. In this study, we establish a novel route choice model, named the paired combinatorial weibit (PCW) model, to capture the route overlap and route-specific perception variance. The PCW model retains a closed-form probability solution, which allows the development of an equivalent mathematical programming (MP) formulation for the PCW-based stochastic user equilibrium (PCW-SUE) model. Specifically, we propose two equivalent MP formulations for modeling the fixed demand (FD) and elastic demand (ED), named PCW-SUE-FD and PCW-SUE-ED, respectively. The PCW-SUE-ED model can address the abovementioned two issues in route choice for the FD scheme, but also can consider the effect level-of-service (LOS) in travel choice for the ED scheme. The equivalency and uniqueness of the PCW-SUE-FD and PCW-SUE-ED models are rigorously proved. In addition, a path-based partial linearization algorithm combined with a self-regulated averaging line search strategy is developed to solve the two SUE models. Numerical results are presented to illustrate the features of the PCW-SUE-FD and PCW-SUE-ED models and applicability of the solution algorithm to a real transportation network.

An Improved Method to Search All Minimal Paths in Networks

Minimal paths (MPs) play an important role in system reliability analysis, such as binary network reliability evaluation and multistate network reliability evaluation. Searching all MPs is an NP-hard problem. The direct-search-based algorithm implements the depth-first search (DFS) mechanism to find all MPs, which is a simple and efficient method. The current efficient DFS algorithm is based on backtracking and linked list by nodes. In this article, we propose several improvements to speed up the current algorithm. First, we find that the relative positional information of the network can be used to avoid visiting some unnecessary nodes. Thus, we improve the algorithm by incorporating the point in the polygon algorithm to avoid visiting nodes surrounded by other nodes in a network. Second, we further improve the current backtracking condition in which the time complexity of implementing each backtrack is reduced. In addition, we find that different search branches from a similar node are independent and can be executed in parallel. A parallel search mechanism is proposed and incorporated to speed up the searching process. Through computational experiments on several benchmark networks and real transportation networks, we demonstrate that the improved algorithm is more efficient than the existing algorithm. The proposed algorithm becomes more advantageous as the size of the network grows.

Multiclass bi-criteria traffic assignment without class-specific variables: An alternative formulation and a subgradient projection algorithm

In this paper, we focus on the multiclass bi-criteria (time and toll) traffic assignment (MBTA) problem. The conventional MBTA model keeps multiple copies of class-specific variables to model user heterogeneity, which puts a great burden on memory storage and computational speed when dealing with real transportation networks. This paper proposes an alternative formulation for the MBTA problem without class-specific variables by exploiting the order information of paths and travelers, i.e., high value of time (VOT) travelers will prefer fast but expensive paths, while low VOT travelers will prefer slow but cheap paths. We prove the equivalence of the alternative formulation to the conventional MBTA model. To solve the alternative formulation with a nondifferentiable convex objective function, a path-based subgradient projection algorithm is developed utilizing the subgradient and available second-order information. We adopt a small network and several large networks to examine the detailed features and the computational performance of the proposed formulation and algorithm, respectively. The results show that the alternative formulation provides the same link flow pattern as that of the conventional MBTA model but uses much fewer variables, which can greatly relieve the burden on computer memory and computational speed in solving real transportation networks.

Analysing inequity in land use and transportation models by genetic algorithm for realistically quantified penetration rate of Advanced Driving System Equipped Vehicles

The continued evolutions in automated driving technologies and their rapid testing on common roads make it necessary to evaluate their impacts on land use and transportation models. It is crucial to quantify the number of advanced driving system-equipped vehicles that are going to be part of transportation networks. On the other hand, the intuitive property of these vehicles to create an induced demand can bring both positive and negative effects on the travel equilibrium costs that create inequity. To cater for the gap of realistic quantification of penetration rate and inequity evaluation on the inclusion of such vehicles; this research crafts a detailed and effective methodology. This research formulates a convex minimization problem as a lower-level part of the bi-level optimization model intending to minimize the travel equilibrium cost for all OD pairs. Also, acts as an assignment of demand to the network following the stochastic user equilibrium approach by using the Frank–Wolfe algorithm. Whereas, the upper level of the model maximizes the production of newly generated demand incorporating inequity constraints. A genetic algorithm is used to solve the multi-objective fitness function yielded from the bi-level optimization model by application of the model on a real transportation network of the city of Genoa, Italy.

Hydrogen station allocation based on equilibrium traffic flow

The deployment of hydrogen refueling stations is vital to the development of fuel cell vehicles. Past studies on the hydrogen refueling stations didn't consider congestion on paths and assumed all flows are on the shortest paths of OD pairs, which is not consistent with travelers' route choice behavior. In this paper, we study flow refueling location allocation based on the user-equilibrium (UE) status, which is consistent with travelers' route choice behavior. We present a flow-capturing location allocation (FCLA) model on UE conditions. We also presented an equilibrium algorithm avoiding cannibalization to solve the model. Based on the transportation network of Anaheim city in California in the US, we compared the results of applying a non-equilibrium method to the FCLA model and applying the equilibrium method to the FCLA model on UE conditions. The results of the two algorithms and models are different. The equilibrium model and algorithm can capture more flow and has a higher flow-capturing ratio. What's more, the equilibrium model and algorithm can fully capture the expected flow in practice since the real transportation network is basically in UE status. The equilibrium model and algorithm presented in this paper is the first study of this kind to locate stations in cities and is useful in the deployment of hydrogen refueling stations.

A link model approach to identify congestion hotspots.

Congestion emerges when high demand peaks put transportation systems under stress. Understanding the interplay between the spatial organization of demand, the route choices of citizens and the underlying infrastructures is thus crucial to locate congestion hotspots and mitigate the delay. Here we develop a model where links are responsible for the processing of vehicles, which can be solved analytically before and after the onset of congestion, and provide insights into the global and local congestion. We apply our method to synthetic and real transportation networks, observing a strong agreement between the analytical solutions and the Monte Carlo simulations, and a reasonable agreement with the travel times observed in 12 cities under congested phase. Our framework can incorporate any type of routing extracted from real trajectory data to provide a more detailed description of congestion phenomena, and could be used to dynamically adapt the capacity of road segments according to the flow of vehicles, or reduce congestion through hotspot pricing.

СРАВНИТЕЛЬНЫЙ АНАЛИЗ МОДЕЛЕЙ ПРОГНОЗИРОВАНИЯ СПРОСА НА ГРУЗОВЫЕ ПЕРЕВОЗКИ

In recent years, there has been a growing awareness of the need for accurate forecasting of freight volumes as part of travel demand assessments. Freight demand forecasting is a complex process that has not been studied to the same extent as passenger demand forecasting. The article compares traditional freight demand forecasting models with tour-based models and discusses the features, advantages, limitations and suitability of models for testing on a real transport network. The aim of the study is to provide an overview of available freight demand forecasting models for inclusion in the transportation planning process

СРАВНИТЕЛЬНЫЙ АНАЛИЗ МОДЕЛЕЙ ПРОГНОЗИРОВАНИЯ СПРОСА НА ГРУЗОВЫЕ ПЕРЕВОЗКИ

The article compares traditional freight demand forecasting models with tour-based models and discusses the features, advantages, limitations and suitability of models for testing on a real transport network. The aim of the study is to provide an overview of available freight demand forecasting models for inclusion in the transportation planning process

A hybrid machine learning-optimization approach to pricing and train formation problem under demand uncertainty

Due to the complexity of pricing in the service industry, it is important to provide an efficient pricing framework for real-life and large-sized applications. To this end, we combined an optimization approach with a regression-based machine learning method to provide a reliable and efficient framework for integrated pricing and train formation problem under hybrid uncertainty. To do so, firstly, a regression-based machine learning model is applied to forecast the ticket price of the passenger railway, and then, the obtained price in is used as the input of a train formation optimization model. Further, in order to deal with the hybrid uncertainty of demand parameters, a robust fuzzy stochastic programming model is proposed. Finally, a real transportation network from the Iran railway is applied to demonstrate the efficiency of the proposed model. The analysis of numerical results indicated that the proposed framework is able to state the optimal price with less complexity in comparison to traditional models.

Navigation methodology for vehicle city route optimal choice

Relevance. The study is a fundamentally new approach to such an extremely important problem as the congestions in large cities. The solution of this global problem is a step in the realization of a smart city concept. Goal. The aim of the study is to create basic elements of technology that can stabilize urban traffic and bring it to a qualitatively new state. To achieve this goal, the following tasks have been formulated: – to create a model of a city transport network in the form of an oriented weighted non-planar multigraph with dynamically loaded arcs; – to activate a city electronic map in the Traffic Management Centre (TMC) which allows tracking each vehicle; – to navigate the time-optimal routes for all those vehicles that request the route; – to implement the work of the software algorithm in real time with constant updating of the route of each tracked vehicle. That will allow monitoring changes in city traffic in real time and making adjustments to the route of each vehicle. Research methods. The research is based on the use of mechanisms for modeling and working with various networks – the graph theory and the A-star algorithm. The latter traces the route on the graph (transport network) between two selected positions of the vehicle graph theory – origin and destination. The heuristic A-star algorithm – a powerful computational method of graph theory has been used in the study. This makes it possible to synchronize vehicles flows and therefore provides a qualitatively new level to the control of urban traffic. The results. The problem of traffic load registration for the city transport network essential for navigating a vehicle route in metropolis has been solved. Traffic data of the real transport network have been reproduced on the city electronic map. Each vehicle received a unique marker consisting of an origin-destination pair and can be tracked on the map. Since each vehicle is under control of the Traffic Management Center (TMC), it is possible to track it along the optimal route, taking an urban traffic dynamic into account. Support is provided via the General Packet Radio Service (GPRS) channel, which allows each driver to receive instructions as to an optimal travel path. Conclusions. The study has proposed a working software module that navigates a time-optimized route on the graph that represents the model of the real transport city network.

Solving a Multi-Class Traffic Assignment Model with Mixed Modes

In comparison to conventional human-driven vehicles (HVs), connected and automated vehicles (CAVs) provide benefits (e.g., reducing travel time and improving safety). However, before the period of fully CAVs appears, there will be a situation in which both HVs and CAVs are present, and the traffic flow pattern may differ from that of a single class (e.g., HV or CAV). In this study, we developed a multi-class traffic assignment problem (TAP) for a transportation network that explicitly considered mixed modes (e.g., HV and CAV). As a link’s travel time is dependent on the degree of mixed flows, each mode required an asymmetric interaction cost function. For TAP, the multi-class user equilibrium (UE) model was used for the route choice model. A route-based variational inequality (VI) formulation was used to represent the multi-class TAP and solve it using the gradient projection (GP) algorithm. It has been demonstrated that the GP algorithm is an effective route-based solution for solving the single-class user equilibrium (UE) problem. However, it has rarely been applied to solving asymmetric UE problems. In this study, the single-class GP algorithm was extended to solve the multi-class TAP. The numerical results indicated the model’s efficacy in capturing the features of the proposed TAP utilizing a set of simple networks and real transportation networks. Additionally, it demonstrated the computational effectiveness of the GP algorithm in solving the multi-class TAP.

Hybrid modeling of collaborative freight transportation planning using agent-based simulation, auction-based mechanisms, and optimization

The sharing economy is a peer-to-peer economic model characterized by people and organizations sharing resources. With the emergence of such economies, an increasing number of logistics providers seek to collaborate and derive benefit from the resultant economic efficiencies, sustainable operations, and network resilience. This study investigates the potential for collaborative planning enabled through a Physical Internet-enabled logistics system in an urban area that acts as a freight transport hub with several e-commerce warehouses. Our collaborative freight transportation planning approach is realized through a three-layer structured hybrid model that includes agent-based simulation, auction mechanism, and optimization. A multi-agent model simulates a complex transportation network, an auction mechanism facilitates allocating transport services to freight requests, and a simulation–optimization technique is used to analyze strategic transportation planning under different objectives. Furthermore, sensitivity analyses and Pareto efficiency experiments are conducted to draw insights regarding the effect of parameter settings and multi-objectives. The computational results demonstrate the efficacy of our developed model and solution approach, tested on a real urban freight transportation network in a major US city.

Planning natural gas networks and storage in emerging countries – an application to Brazil

Some emerging countries, such as Brazil, have large remaining natural gas resources but relatively poor infrastructure to monetize it. When most of the natural gas extraction derives from associated gas, this results in high reinjection rates in production fields combined with fuel imports also to deal with an increasingly variable demand. This study test the hypothesis that modeling the natural gas transportation network expansion with Underground gas Storage (UGS) is crucial, as UGS can reduce transportation costs by better fitting natural gas supply and demand. Without UGS chances are that network expansion will be based in oversized pipelines, or pipelines often challenged by peaking demands. Therefore, this study emulated a real natural gas transport network in a thermo-hydraulic model, aiming at diagnosing its bottlenecks mainly caused by demand intermittency, and pointing out infrastructure solutions. Findings indicated the design of UGS associated with new pipelines as a problem-solver for network bottlenecks, under a least-cost approach. This option reduced idleness and lowered gas transmission costs by 60%. In addition, it increased the network operation reliability and created a virtuous cycle, where a better planning reduces the gas tariffs and spur infrastructure expansion by raising the fuel competitiveness.

Are current tsunami evacuation approaches safe enough?

Developing an effective tsunami evacuation plan is essential for disaster risk reduction in coastal regions. To develop effective tsunami evacuation plans, real transportation network, interaction among evacuees, and uncertainties associated with future tsunami events need to be considered in a holistic manner. This study aims to develop such an integrated tsunami evacuation approach using agent-based evacuation simulation and advanced stochastic tsunami hazard assessment. As a case study, a urban area in Padang, Indonesia, threatened by tsunamis from the Mentawai–Sunda subduction zone, is adopted. The uncertainty of the tsunami hazard is taken into account by generating 900 stochastic tsunami inundation maps for three earthquake magnitudes, i.e. 8.5, 8.75, and 9.0. A simplified evacuation approach considering the evacuees moving directly to evacuation areas (defined a priori) is compared with two more rigorous agent-based modeling approaches: (a) a two-destination-point tsunami evacuation plan developed by the local government and (b) a multiple-destination-point plan developed in this study. The improved agent-based stochastic tsunami evacuation framework with multiple destinations takes advantage of the extensive tsunami hazard analyses to define safe areas in a dynamic manner and is capable of capturing the uncertainty of future tsunami risk in coastal areas. In contrast, the results clearly show that the simplified approach significantly underestimates the evacuation time, and the existing tsunami evacuation routes identified by local authorities may be insufficient to save lives.

Percolation of heterogeneous flows uncovers the bottlenecks of infrastructure networks

Whether it be the passengers’ mobility demand in transportation systems, or the consumers’ energy demand in power grids, the primary purpose of many infrastructure networks is to best serve this flow demand. In reality, the volume of flow demand fluctuates unevenly across complex networks while simultaneously being hindered by some form of congestion or overload. Nevertheless, there is little known about how the heterogeneity of flow demand influences the network flow dynamics under congestion. To explore this, we introduce a percolation-based network analysis framework underpinned by flow heterogeneity. Thereby, we theoretically identify bottleneck links with guaranteed decisive impact on how flows are passed through the network. The effectiveness of the framework is demonstrated on large-scale real transportation networks, where mitigating the congestion on a small fraction of the links identified as bottlenecks results in a significant network improvement.

An Integrated Inventory and Distribution Problem for Alternative Fuel: a Matheuristic Approach

Due to the limited driving range of alternative fuel (AF) vehicles and their immature refueling infrastructure, a successful transition to the era of AF vehicles necessitates ensuring stable supply and management of AF in the refueling network. This paper proposes a new mathematical framework to solve an AF inventory and distribution problem in which an AF provider manages and operates the AF refueling network to meet all AF demand in a given time horizon. As a solution method, we present a mixed-integer programming (MIP) model that minimises the sum of AF service, inventory holding, and distribution costs. Furthermore, a matheuristic algorithm hybridising an MIP model and an adaptive large neighbourhood search algorithm is designed to solve practical problems of real transportation networks. The proposed matheuristic algorithm is validated with an application to small-size instances and is applied to six states in the USA with real traffic flows. [Received: 18 September 2019; Accepted: 6 November 2020]