Congestion Level Research Articles

EcoMRL: Deep reinforcement learning-based traffic signal control for urban air quality

As greenhouse gas (GHG) emissions from road networks significantly impact the urban environment, there has been a growing interest in integrating environmental factors into traffic controls. Traffic signal control (TSC) is an effective method to reduce air pollution by managing traffic congestion. Accordingly, several recent studies have applied artificial intelligence to TSC to improve traffic efficiency. Despite these efforts, there are a few limitations on the existing studies. First of all, a number of recent studies focus only on traffic performance rather than environmental effects. However, in some cases, the objective function related to traffic efficiency can have an adverse effect on the environment. Furthermore, the variable nature of emissions based on traffic demand is rarely considered. To address these issues, we propose a multi-agent deep reinforcement learning-based TSC to minimize GHGs in traffic networks. To this end, we first identify vehicle’s stop and delay as key factors in traffic efficiency that critically influence air quality. Subsequently, we integrate these factors into our model’s reward function. To validate the proposed model, we conduct experiments under various demand scenarios featuring different traffic volume levels. The results show that the proposed model outperforms comparative control methods in two aspects: traffic performance and environmental sustainability. Especially at a high congestion level, the proposed model reduces average delay time by 11.9–16.1% and increases total throughput by 11.5%. Additionally, it significantly lowers total CO2 emissions by up to 8.9% and average emissions by up to 18.3%.

A Learnheuristic Algorithm Based on Thompson Sampling for the Heterogeneous and Dynamic Team Orienteering Problem

The team orienteering problem (TOP) is a well-studied optimization challenge in the field of Operations Research, where multiple vehicles aim to maximize the total collected rewards within a given time limit by visiting a subset of nodes in a network. With the goal of including dynamic and uncertain conditions inherent in real-world transportation scenarios, we introduce a novel dynamic variant of the TOP that considers real-time changes in environmental conditions affecting reward acquisition at each node. Specifically, we model the dynamic nature of environmental factors—such as traffic congestion, weather conditions, and battery level of each vehicle—to reflect their impact on the probability of obtaining the reward when visiting each type of node in a heterogeneous network. To address this problem, a learnheuristic optimization framework is proposed. It combines a metaheuristic algorithm with Thompson sampling to make informed decisions in dynamic environments. Furthermore, we conduct empirical experiments to assess the impact of varying reward probabilities on resource allocation and route planning within the context of this dynamic TOP, where nodes might offer a different reward behavior depending upon the environmental conditions. Our numerical results indicate that the proposed learnheuristic algorithm outperforms static approaches, achieving up to 25% better performance in highly dynamic scenarios. Our findings highlight the effectiveness of our approach in adapting to dynamic conditions and optimizing decision-making processes in transportation systems.

Aplicación de la minería de datos espacial basado en técnicas de agrupamiento para detectar los niveles de congestionamiento del tráfico vehicular en la ciudad de Trujillo

Aplicación de la minería de datos espacial basado en técnicas de agrupamiento para detectar los niveles de congestionamiento del tráfico vehicular en la ciudad de Trujillo

An overview of dynamic pricing toll roads in the United States: Pricing algorithms, operation strategies, equity concerns, and funding mechanism

Toll roads with fully dynamic pricing schemes could play a key role in addressing the economic and environmental tolls imposed by increasing congestion levels in major metropolitan areas. Transportation professionals usually argue that they could provide additional revenue streams to accelerate the delivery of much-needed highway projects and serve as a tool for travel demand management. However, the number of dynamic pricing toll roads only increased recently in the U.S., and their performance has not been empirically evaluated systematically. This study provides a comprehensive review of U.S. dynamic pricing toll roads. Pricing is reactive to demand, and additional travel demand management strategies such as incentives for high occupancy vehicles, clean vehicles, and transit are commonly included. Both conventional project delivery methods and private–public partnerships have been used for such projects. Financial statements from a few P3 dynamic toll road projects suggest they are taking full advantage of TIFIA loans and are performing well during their short history. Equity issues are commonly raised, and many projects address them specifically through subsidies to vulnerable groups or re-investment in the community. Further quantitative analyses are needed to test if dynamic toll roads are an efficient tool to address the financial, economic, and environmental challenges major metropolitan areas face.

Integrated optimization of multi-machine scheduling and berth allocation in automated ore terminals

As water transportation undergoes a new technological revolution, automated terminal technology has become a key issue in reducing costs and improving efficiency. However, the operation theory limitation makes it problematic for machines to cooperate effectively and operate profitably, which is notably salient at ore terminals with their unique structure. This study focuses on the integrated scheduling of a new automated ore terminal. It comprises three main processes: (a) berth allocation, (b) splitter scheduling, and (c) feeder scheduling. We attempt to tackle this issue by developing a mixed-integer linear programming (MILP) model to minimize the vessels’ maximum departure time. Considering the complexity of the MILP model, we propose a logic-based Benders decomposition algorithm to solve the efficiency problem. The results indicate that the algorithm can solve realistically sized problem instances. The main factors affecting the computation time are the congestion level, total cargo types, and vessel service time. The integrated scheduling methods can effectively solve the actual challenges faced by terminals and provide suggestions and references for terminal expansion planning.

Transaction Analysis for Operational Optimization in Transjakarta Public Transportation System

This study presents a comprehensive exploratory data analysis (EDA) of transaction data within the Transjakarta public transportation system, focusing on routes, passenger flow, and congestion patterns. Leveraging tap-in and tap-out transaction records collected through payment cards, the research aims to provide actionable insights for optimizing operational efficiency and enhancing service quality. By analyzing transaction data, the study identifies busy routes, peak hours, and passenger distribution across different modes of transportation, including big bus (BRT), medium and big bus (non-BRT), and mini-bus (Mikrotrans). Moreover, the research investigates correlations between transaction volume and traffic congestion levels, enabling the identification of areas prone to delays and bottlenecks. The findings of this analysis offer valuable information for route planning, scheduling adjustments, and infrastructure improvements within the Transjakarta system. By optimizing resources based on data-driven insights, the study seeks to contribute to the enhancement of public transportation services, promoting accessibility, reliability, and sustainability in urban mobility.

Relationship between the Significant Solutions of Static Traffic Assignment Problems for Mixed Traffic Flow of Connected and Automated Vehicles and Human-Driven Vehicles

Connected and automated vehicles can reduce the traffic congestion level of the entire network through platoon-driving technologies compared to human-driven vehicles. One promising approach to enhancing platoon-driving technology’s efficiency is deploying dedicated lanes or roads for connected and automated vehicles. Since asymmetric interactions between different vehicle types increase road congestion, it is necessary to distinguish routes for efficient traffic management. However, the traditional traffic assignment problem, which uses only user equilibrium as a constraint with no difference in travel time between users, could not be proposed as a globally optimal solution because it generates an infinite number of locally optimal solutions. Recent studies have attempted to overcome the limitations by considering the sum of system-wide travel times as an additional constraint. Their research sought to help propose optimal deployment strategies through the lowest total travel time solution (best-case) or design robust transport planning strategies through the highest total travel time solution (worst-case). However, past studies have not focused on the possibility of the best/worst case appearing in reality. This study focused on the relationship between the two solutions pointed out in past studies and traffic patterns likely to appear in reality. This study interprets the Karush–Kun–Tucker condition of the static traffic assignment problem, considering the asymmetric interaction, and proposes a solution algorithm using discrete dynamics. The proposed algorithm extends the most widely used method in transportation planning research, which can overcome the limitations of asymmetric interaction problems through simple variations. The proposed algorithm can reliably derive two solutions, and entropy theory shows that both solutions are unlikely to appear in reality without additional policies such as dedicated lanes or roads.

Operative Benefits of Residential Battery Storage for Decarbonizing Energy Systems: A German Case Study

The reduction in PV prices and interest in energy independence accelerate the adoption of residential battery storage. This storage can support various functions of an energy system undergoing decarbonization. In this work, operative benefits of storage from the system perspective, namely, generation cost reduction and congestion mitigation, are investigated. Germany is chosen as a case study due to its strong reliance on variable renewable energy. For the analysis, an economic dispatch model with a high spatial resolution is coupled with a pan-European transmission grid model. It is shown that the system’s generation costs are highest when the assets are used only to maximize PV self-consumption, and the costs are lowest when the storage also reacts to the market dynamics. This amounts to a 6% cost reduction. Both operation strategies result in an equal level of grid congestion and infrastructure loading. This is improved with a strategy that accounts for regional peak reduction as a secondary objective. The high congestion level emphasizes that grid expansion needs to keep pace with the generation and electrification expansion necessary to decarbonize other sectors. Lastly, policymakers should enable multipurpose utilization, e.g., via the introduction of market-oriented retail electricity prices with intervention options for grid operators.

A cellular automaton model for mixed traffic flow considering the size of CAV platoon

In the environment of intelligent connected vehicles, the mixed traffic flow consisting of human-driven vehicles (HDVs), connected and automated vehicles (CAVs), and their platoons will coexist for a long time. Exploring and discovering the operational rules of mixed traffic flow is the foundation for its management and control. This study delves into the distinct characteristics of various vehicles within a mixed traffic environment, particularly focusing on HDVs, CAVs, and CAV platoons. Drawing on the psychological predisposition of HDVs to maintain a prudent following distance, we introduce a discrete motion safety distance model to establish the longitudinal movement rule for HDVs. Furthermore, a longitudinal movement rule for CAVs is established based on their ability to keep a constant headway using Adaptive Cruise Control (ACC) and Cooperative Adaptive Cruise Control (CACC), alongside meeting specific speed and spacing criteria for platoon movement. On this basis, the paper proposes a cellular automaton model for single-lane mixed traffic flow that accounts for CAV platoon length and the decision-time disparity between HDVs and CAVs. The simulation results show that an increased proportion of CAVs can enhance the traffic capacity, elevate average speed, extend average platoon length, and reduce the congestion ratio and CACC degradation rate. Notably, in scenarios where the CAV proportion p > 0.7, the flow-density graph shows a trapezoidal shape, with the majority of CAVs in the traffic flow synchronizing their movements in CACC mode. Even when vehicular density surpasses the optimum threshold, the traffic flow maintains high-speed operation until reaching a critical congestion point, whereupon velocities precipitously decline. Through sensitivity analysis aimed at understanding the impact of maximum platoon length, it has been observed that, for CAV proportions below 0.7, variations in maximum platoon length negligibly impact traffic capacity, velocity, CACC degradation, and congestion levels. Only in scenarios dominated by CAVs does extending platoon length restrictions significantly enhance performance. However, as the maximum platoon length increases, the gains in traffic capacity, speed, CACC degradation rate, and congestion ratio diminish. When the platoon length of CACC exceeds a certain threshold, there will not be a significant improvement in traffic capacity or speed.

Enhancing Make-to-Order Manufacturing Agility: When Flexible Capacity Meets Dynamic Pricing

The rise of online marketplaces has raised customer expectations regarding customization and lead time. It poses significant challenges to manufacturing firms and prompts a move from make-to-stock to a more flexible make-to-order system. Compared to make-to-stock settings, make-to-order systems cannot smooth fluctuations in demand using available stock. While viewing dynamic pricing as a useful strategy to balance supply with demand, many manufacturing firms can also create capacity flexibility. In that scenario, system costs could be cut by managing capacity and demand simultaneously. In this paper, we consider a make-to-order production environment with base and surge capacity as well as the ability to adjust product pricing. Our main focus is on operational decision-making, assuming that the base capacity and surge capacity are fixed, but activating the surge capacity incurs a setup cost. Initially, we propose a stochastic control model to reflect this complex decision problem. However, our initial model leads to an intractable dynamic programming problem. To overcome this, we convert the problem to a more tractable diffusion control problem. This approach helps to reveal the conditions under which utilizing flexible capacity is more advantageous than relying solely on fixed capacity. When flexible capacity is advantageous, we provide a solution to the diffusion control problem that can guide optimal capacity and price adjustments. We discover an interesting interplay between capacity adjustment and dynamic pricing. In particular, we find that the price, which aims at reducing congestion, may not monotonically increase with the congestion level when capacity adjustments incur a fixed cost.

Street Crowdedness and Urban Congestion Level by MAC and GPS Data

Wi-Fi is one of the emerging technologies today. Wi-Fi refers to MAC (Media Access Control) address data. A MAC address represents a device number. In other words, it represents the number of people. The system uses MAC address data to determine road traffic congestion and land use, detects the number of terminals using a Wi-Fi scanner, and analyzes vehicle speed using GPS. The purpose of this research is to understand city-level services from traffic services and road congestion status and degree of congestion using Wi-Fi data (MAC addresses). This method was used for this study by providing a Wi-Fi scanner and a GPS scanner inside the bus. A Wi-Fi scanner that collects MAC address data and GPS for calculating bus speed. The congestion hypothesis in this study is that the bus is slow, and the total number of MAC addresses is high at the same time, or vice versa. The results show that the maximum correlation value between speed and total number of MAC addresses is -0.629. Negative correlation leads to slower bus speeds, higher total MAC addresses, and different. The numbers indicate the congested sections around the bus route during the bus operating hours. Several parameters such as travel time, travel speed and standard deviation/variance were analyzed for congestion level. Analyze aspects that link road conditions to good or bad service. Congestion levels can be determined based on RI, which is derived from the standard deviation and average speed of each road link.

Transport policies and their everyday impact on mobilities of older adults in Bengaluru

Background and objectives: Cities in the Global South are simultaneously witnessing wide-scale population ageing and rapid urbanisation. There is overwhelming pressure on policy makers to rapidly envision and provision age-friendly transport environments. In Indian cities, later-age mobilities face unique risks due to traffic congestion and inequities. However, the policy and transport research has treated congestion, equity and age-friendly mobilities as unrelated issues. This article goes a step ahead to understand how the policies to mitigate congestion can impact the provisioning of age-friendly infrastructure. Research design and methods: Bengaluru, once termed the ‘pensioners’ paradise’, is known for its ‘alarming’ levels of traffic congestion (Pucher et al, 2005; Pojani and Stead, 2017). We review Bengaluru’s transport policies and peer-reviewed articles concerning these policies to understand the positioning of age-friendly transport within this context. Results: Our findings recognise a shift from sufficientarianism to utilitarian orientations in Bengaluru’s policy making on easing congestion, which narrows the options for later-age transport users in terms of access and affordability. Some of the features not unique to Bengaluru – such as top-down planning, institutional fragmentation and distancing of users from policy making – have resulted in buses being inaccessible for passengers with mobility issues, metro rail being unaffordable for those on low incomes, lack of safe modes of transport and safe access to streets. Discussion and implications: The article is an initial foray into the ambiguity in the philosophical underpinning of transport policy making, which prioritises solving congestion ahead of providing age-friendly transport. We highlight specific challenges the cities of the Global South experience as they envision age-friendly transport policies amid the ascendance of congestion-solving policies.

Rethinking LEO constellations routing

SummaryThis study investigates the unsplittable multicommodity flow (UMCF) as a routing algorithm for LEO constellations. Usually, LEO routing schemes enable the Floyd–Warshall algorithm (shortest path) to minimize the end‐to‐end latency of the flows crossing the constellation. We propose to solve the UMCF problem associated with the system as a solution for routing over LEO. We use a heuristic algorithm based on randomized rounding known in the optimization literature to efficiently solve the UMCF problem. Furthermore, we explore the impact of choosing the first/last hop before entering/exiting the constellation. Using network simulation over Telesat constellation, we show that UMCF maximizes the end‐to‐end links usage, providing better routing while minimizing the delay and the congestion level, which is an issue today over new megaconstellations.

A Multi-Agent Reinforcement Learning-Based Grant-Free Random Access Protocol for mMTC Massive MIMO Networks

The expected huge number of connected devices in Internet of Things (IoT) applications characterizes the massive machine-type communication (mMTC) scenario, one prominent use case of beyond fifth-generation (B5G) systems. To meet mMTC connectivity requirements, grant-free (GF) random access (RA) protocols are seen as a promising solution due to the small amount of data that MTC devices usually transmit. In this paper, we propose a GF RA protocol based on a multi-agent reinforcement learning approach, applied to aid IoT devices in selecting the least congested RA pilots. The rewards obtained by the devices in collision cases resemble the congestion level of the chosen pilot. To enable the operation of the proposed method in a realistic B5G network scenario and aiming to reduce signaling overheads and centralized processing, the rewards in our proposed method are computed by the devices taking advantage of a large number of base station antennas. Numerical results demonstrate the superior performance of the proposed method in terms of latency, network throughput, and per-device throughput compared with other protocols.

Traffic Congestion Analysis of Asaba Road Using Volume to Capacity Ratio and Speed Performance Index

With Nnebisi Road acting as a case study, this study focuses on the investigation of traffic congestion on Asaba Road. The study aims to examine the relationship between vehicle volume and road capacity, enabling transportation authorities to assess congestion levels and make informed decisions to alleviate traffic issues. Peak traffic counts were conducted, revealing the highest traffic flow on specific road sections on different days. The Ibusa – Stadium junction saw the highest volume on Monday, reaching 3,056 vehicles, while Ibusa – Ofili-Ilukwu junction peaked at 3,128 vehicles, also on Monday, and Ibusa – Ogbeogonogo junction hit 3,151 vehicles on Tuesday. Speed performance index varied across junctions and days, generally exceeding 60%. For instance, Ibusa – Stadium Junction ranged from 69.57% to 71.82% from Monday to Friday, Ibusa – Ofili-Ilukwu junction showed values between 63.59% and 68.95%, and Ibusa – Ogbeogonogo Junction ranged from 61.57% to 70.11%. These findings suggest a better road traffic state with higher average speeds. Moreover, the volume to capacity ratio, indicates the level of congestion and delays experienced. The ratio values along the studied road sections ranged from 0.61 to 0.70. These ratios, along with the speed performance index, classify the level of service as class B, indicating moderate congestion and delays with slightly longer travel times, yet still acceptable to users.

A combined short- and medium-term traffic flow prediction method for proactive traffic control at expressway toll stations

Traffic flow prediction can support proactive control strategies to alleviate queues and delays at expressway toll stations. The prediction should be enough accurate and advanced that leaves sufficient preparation time to implement control measures for upcoming congestions. The accuracy of prediction and the lead time for accurate prediction are contradictory. To balance both aspects, a control-oriented combined prediction method was proposed in this study. This method automatically tuned prediction horizons based on predicted level of congestion. It used long short-term memory (LSTM) neural networks to learn the spatial–temporal characteristics of historical traffic flow and predict short-term and medium-term traffic flow at toll stations. Then, the level of congestion was determined for graded congestion warning and dynamic tuning of prediction horizons. To validate the proposed method, the Ganggou toll station was selected as the study site. The proposed integrated model can facilitate reliable proactive control measures at expressway toll stations.

Spatial system perspective of understanding “fuel-sensitive routes” using regional-scale case studies

Eco-routing is widely researched as a promising way to reduce traffic-related environmental externalities. Past studies indicate that fuel-sensitive eco-routing trades the fastest routes with shorter, slower urban roads to take advantage of the “optimal fuel economy”. However, past studies focusing on individual routes often fail to depict the spatial redistribution of traffic being closer to the residential areas when eco-routing is adopted at a large scale. This research conducts city-scale fuel-sensitive routing simulations for two large metropolitan areas: the San Francisco Bay Area case and the Tokyo Area case. Combining agent-based simulations with data-driven fuel model, it is revealed that fuel-sensitive routing exerts heavier traffic on the local streets and increases residents’ direct exposure to traffic. Furthermore, by comparing different sub-areas, the varied degrees of traffic redistribution are found to be mainly dependent on the availability of major roads as alternative routes, and existing traffic congestion levels in an area.

Analysis of the Effect of Side Obstacles on Performance on the Z.A. Pagaralam Road Section

Development in Bandar Lampung is increasing rapidly, causing an increase in transportation development. The development of transportation in the city of Bandar Lampung is currently still experiencing severe problems, so traffic hours often occur. The high level of congestion on the highway is caused by several things, including the unbalanced road capacity and vehicle volume and the presence of side obstacles, which add to congestion on the Z.A. Pagaralam road section. Apart from being a public road, Pagaralam is also located in one of the busiest economic centers in the city of Bandar Lampung. Along this stretch of road are campuses, schools, shops, pharmacies, shopping centers, and street vendors, and it is also a transfer point for passengers moving from one public transport to another, causing many side obstacles to occur on the road. To overcome this, it is necessary to analyze the influence of side obstacles on traffic performance. Data analysis was carried out using the PKJI 2014 method. The analysis of traffic volume obtained a value of 4747.75 cur/hour. The analysis results of the highest side obstacles were obtained on Tuesday at 12.00 – 14.00 segment 1 amounting to 838.90 incident weights. Keywords: Traffic Volume, Side Obstacles, PKJI 2014.

Probabilistic modeling and simulation of drivers’ lane selection and its impact on highway traffic using fuzzy-based decision-making

Traffic on highways has increased significantly in the past few years. Consequently, this has caused delays for the drivers in reaching their final destination and increased the highway’s congestion level. Many options have been proposed to ease these issues. In this paper, a model of the highway drivers’ population was built based on several factors, including the behavioral patterns of the drivers, like drivers’ time flexibility to reach the destination, their carpool eligibility, and their tolerance to pay the toll price, in addition to the traffic information from the system. A fuzzy logic decision-making model is presented to emulate how drivers would choose the lane to use based on the aforementioned factors and the current congestion levels of all the lanes on the highway. The presented model, along with the simulation results from applying the model to different simulation scenarios, show the usefulness of such a model in predicting an optimal toll value. Such optimal value would reduce congestion on the highway at one end while maximizing the revenue for the toll company.

Enhancing Transshipment Decision-making in Transportation Companies through Computational Intelligence: A Bayesian Game Model with Data-Driven Simulations at Guangzhou Port

The challenge of port congestion significantly impedes the efficiency of global trade flows and the economic vitality of hinterland regions. Addressing this issue, our study harnesses computational intelligence to construct a sophisticated Bayesian game model between the government and multiple transportation companies, where the transshipment costs, considered as private information to the companies, play a pivotal role. This research integrates data analytics and machine learning techniques to analyze the strategic decision-making processes of transportation companies, which decide on transshipment based on a critical cost threshold influenced by government subsidies and a probabilistic assessment of transshipment costs. Utilizing backward induction, the study outlines how the government can leverage computational models to devise an optimal subsidy strategy for transshipment, taking into account the anticipated responses of the transportation companies. The Bayesian Nash equilibrium identified through our model suggests that companies with costs below a predefined threshold are incentivized by government subsidies to opt for transshipment. This conclusion is further validated through evolutionary game theory analysis, enriched by data-driven simulations. Employing real-world data from Guangzhou Port, we conducted extensive computational simulations to quantify the impact of transshipment subsidies. The findings reveal a substantial alleviation in port congestion, with a 33.7% reduction in congestion levels and a 35% decrease in congestion-related costs, alongside a notable 1.9% increase in government revenue. These simulations, powered by advanced computational algorithms and data analytics, not only underscore the effectiveness of informed subsidy strategies in mitigating port congestion but also demonstrate the potential of integrating computational approaches in logistical and transportation decision-making. This study contributes a novel computational framework to the logistics and transportation literature, offering practical insights for policymakers to tackle the enduring problem of port congestion through data-driven strategies.