Traffic Allocation Research Articles

Social movements boosted online orders for US Black-owned restaurants after the murder of George Floyd.

We use the rise of Black Lives Matter and the sentiment of racial sympathy to examine the interplay between the social movement and citizens' sympathetic actions in supporting Black people. Using detailed food order flow information from one of the largest online food delivery platforms in the USA, we find that the total number of food orders from Black-owned restaurants increased by 39% relative to nearby non-Black-owned restaurants in the 140 days following the murder of George Floyd on the basis of a difference-in-difference model. The platform company's strategic traffic allocation acted as an accelerator, enhancing the sympathetic responses of individuals, but it did not drive the entire surge in food orders. Protests resulting in severe injuries and those linked to demands for defunding the police diminished the positive sympathetic responses, highlighting a potential risk associated with protests. Our study provides large-scale, micro-level evidence that social movements and increased sympathy can foster collective actions to support marginalized communities.

Circling Reduction Algorithm for Cloud Edge Traffic Allocation Under the 95th Percentile Billing

Circling Reduction Algorithm for Cloud Edge Traffic Allocation Under the 95th Percentile Billing

Neural quantile optimization for edge–cloud networking

We seek the best traffic allocation scheme for the edge–cloud networking subject to SD-WAN architecture and burstable billing. First, we formulate a family of quantile-based integer programming problems for a fixed network topology with random parameters describing the traffic demands. Then, to overcome the difficulty caused by the discrete feature, we generalize the Gumbel-softmax reparameterization method to induce an unconstrained continuous optimization problem as a regularized continuation of the discrete problem. Finally, we introduce the Gumbel-softmax sampling neural network to solve optimization problems via unsupervised learning. The neural network structure reflects the edge–cloud networking topology and is trained to minimize the expectation of the cost function for unconstrained continuous optimization problems. The trained network works as an efficient traffic allocation scheme sampler, outperforming the random strategy in feasibility and cost value. Besides testing the quality of the output allocation scheme, we examine the generalization property of the network by increasing the time steps and the number of users. We also feed the solution to existing integer optimization solvers as initial conditions and verify the warm-starts can accelerate the short-time iteration process. The framework is general, and the decoupled feature of the random neural networks is adequate for practical implementations.

Identifying a selection mechanism of distribution channel for the supply chain: The barriers to the application of web 3.0

The application of Web 3.0 will bring unprecedented changes to the supply chain, where many industries are seeing significant cost reductions. However, due to the requirement of stability for supply chain, sometimes the total cost of supply chain cannot decrease effectively under Web 2.0, which may hinder the application of Web 3.0. Following the application from Web 2.0 to Web 3.0, this paper discusses the selection of distribution channel whose cost is under user stickiness and the degree of digitalization. The system optimum model, which is commonly used in dynamic traffic allocation, is applied to solve the distribution channel selection problem. The results implies that when the degree of digitalization is low, the distribution channel cost first decreases and then increases as user stickiness increases. This trend of channel cost illustrates the price induction on user stickiness initially, and gradually adjust the price to be higher when user stickiness makes channel transformation costly. Customers with unclear user stickiness perceptions and weak channel control can easily be trapped in the transition from Web 2.0 to Web 3.0. Distributors will also be forced to accept high product costs due to the price induction from product distribution channels.

A robust optimization approach for placement of applications in edge computing considering latency uncertainty

Edge computing brings computing and storage resources close to end-users to support new applications and services that require low network latency. It is currently used in a wide range of industries, from industrial automation and augmented reality, to smart cities and connected vehicles, where low latency, data privacy, and real-time processing are critical requirements. The latency of accessing applications in edge computing must be consistently below a threshold of a few tens of milliseconds to maintain an acceptable experience for end-users. However, the latency between users and applications can vary considerably depending on the network load and mode of wireless access. An application provider must be able to guarantee that requests are served in a timely manner by their application instances hosted in the edge despite such latency variations. This article focuses on the placement and traffic allocation problem faced by application providers in determining where to place application instances on edge nodes such that requests are served within a certain deadline. It proposes novel formulations based on robust optimization to provide optimal plans that protect against latency variations in a configurable number of network links. The robust formulations are based on two different types of polyhedral uncertainty sets that offer different levels of protection against variations in latency. Extensive simulations show that our robust models are able to keep the number of chosen edge nodes low while reducing the number of latency violations as compared to a deterministic optimization model that only considers the average latency of network links.

Potential game in parallel transport network with symmetric externalities

The paper considers a model of a transport system with parallel channels and BPR delay functions with symmetric linear externalities. We consider the case where the impact of channel loads to the delay is pairwise symmetric. For this case, it is proved that the game of traffic allocation among channels is potential, and the price of anarchy is limited by the value $\frac{4}{3}$.

Viewer Traffic Allocation for Small Creator Development: Experimental Evidence from Short-Video Platforms

Viewer Traffic Allocation for Small Creator Development: Experimental Evidence from Short-Video Platforms

Huawei Cloud Adopts Operations Research for Live Streaming Services to Save Network Bandwidth Cost: The GSCO System

The rapid evolution of cloud computing technologies has instigated a paradigm shift across various traditional industries, with the live streaming sector standing as a compelling exemplification of this transformation. Huawei Cloud, which has become an influential player in the business-to-business live streaming arena, with its services spanning over 60 countries since 2020, is at the forefront of this shift. Amid the flourishing live streaming market, Huawei Cloud faces the dual challenge of satisfying the escalating demand, while managing the mounting operational costs, predominantly associated with the network bandwidth. To offer premium services while minimizing the bandwidth cost, we developed a dynamic traffic allocation system called GSCO. This system was engineered using an array of operations research methodologies such as continuous optimization, integer programming, graph theory, scheduling, and network-flow problem solving, along with state-of-the-art machine learning algorithms. The GSCO system has been proven highly effective in cost optimization, reducing network bandwidth expenses by about 30% and leading to savings exceeding $49.6 million from Q1 2020 to Q3 2022. In addition, it has significantly bolstered Huawei Cloud’s market share, amplifying peak bandwidth from an initial 1.5 terabits per second (Tbps) to a substantial 16 Tbps.

Fasys: Visible-Light-Based Communication and Positioning Services towards Smart Cities.

Visible-light-based transmission application plays an important role in various types of sensor services for the Internet of Things (IoTs). However, in big data scenarios, current visible-light-based systems cannot achieve concurrent high-speed communication, low-speed communication, and positioning. Therefore, in this article, we propose a smart visible-light-based fusion applications system, named Fasys, to solve the above problem for the big data traffic with heterogeneity. Specifically, for low-speed data services, we propose a novel linear block coding and bit interleaving mechanism, which enhances the LED positioning accuracy and recovers the lost data bits in the interframe gap (IFG). For high-speed data services with traffic possessing burstiness, an elegant statistical reliability analysis framework in regard to latency is proposed based on martingale theory. The backlog martingale process is constructed. Leveraging stopping time theory, a tight upper bound of unreliability is obtained. An arrival abstraction and traffic allocation scheme is designed, which contributes to decouple the reliability requirement as the maximum supportable arrival load. Finally, we implement our Fasys system, and extensive experimental results show that our system can achieve consistent high-precision positioning and low-BER data communication for low-speed data services. And the proposed martingale-based traffic allocation scheme can achieve the provisioning of reliability in regard to the latency for high-speed data services.

Research on the Collaborative Optimization of the Power Distribution Network and Traffic Network Based on Dynamic Traffic Allocation

With the increasing penetration rate of electric vehicles, the spatiotemporal coupling relationship between the power distribution network and traffic network is stronger than ever before. Under the dynamic wireless charging mode, traffic jam charging is introduced and the dynamic loading process of traffic flow is described using a cellular transmission model. The charging load is related to traffic flow and serves as a bond between the power distribution network and traffic network. The traffic flow achieves balanced allocation under dynamic user equilibrium conditions, and cooperatively optimizes the power flow of the power distribution network in conjunction with charging loads. Numerical analysis shows that this model can accurately depict the congestion situation during peak travel periods, and alleviate traffic congestion and distribution network voltage out of range.

The Traffic Flow Prediction of Newly-Constructed Expresseway Based on ETC Database

The traffic flow prediction of newly-constructed expressway can provide reference for highway capacity design and highway toll standard formulation, but the traffic flow prediction of newly-constructed expressway is always considered a complicated project. With the wide application of electronic toll collection system, it is possible to effectively predict the traffic flow of newly-constructed expressways. In this paper, We propose a new highway traffic flow prediction method based on ETC database. ETC database can be used to extract OD matrix of expressway network, and the dynamic traffic allocation model is applied to predict the traffic flow of newly-constructed expressway. The experimental results show that the method is feasible and effective.

Dynamic Traffic Allocation Model Considering the Effects of Vehicle Emission Diffusion

Vehicle exhausts diffuse into roadside crowd breathing zones, thereby jeopardising human health. This study applies dynamic traffic distribution theories to comprehensively consider the impact of vehicle emission diffusion. The results provide a theoretical basis for improving the diffusion of urban traffic pollution to benefit the surrounding environment for roadside crowds. Firstly, a multi-vehicle cellular transport model that is suitable for analysing dynamic traffic distribution is constructed considering the distinct emission factors of various types of vehicles. Secondly, a multi-vehicle emission model is established to consider a range of driving conditions. Then, the concept of roadside crowd exposure risk is introduced, and we describe a method for calculating the total amounts of pollutants emitted by vehicles and inhaled by roadside crowds. The impact of vehicle emission diffusion is comprehensively discussed in terms of vehicle emissions and roadside crowd exposure risk. A generalised impedance function considering the influence of vehicle exhaust emission diffusion is also established based on the weighted average of actual vehicle travel time, multi-model emissions and roadside crowd exposure risk. Finally, this generalised impedance function is integrated into the dynamic optimal user allocation model, and a dynamic traffic allocation model considering the influence of vehicle emission control is developed.

Q-Learning-Based Energy-Efficient Network Planning in IP-Over-EON

During network planning phase, optimal network planning implemented through efficient resource allocation and static traffic demand provisioning in IP-over-elastic optical network (IP-over-EON) is significantly challenging compared with the fixed-grid wavelength division multiplexing (WDM) network due to increased flexibility in IP-over-EON. Mathematical programming based optimization models used for this purpose may not provide solution for large networks due to large computational complexity. In this regard, a greedy heuristic may be used that intuitively selects traffic elements in sequence from static traffic demand matrix and provisions the traffic elements after necessary resource allocation. However, in general, such greedy heuristics offer suboptimal solutions, since appropriate traffic sequence offering the optimal performance is rarely selected. In this regard, we propose a reinforcement learning technique (in particular a Q-learning method), combined with an auxiliary graph (AG)-based energy efficient greedy method to be used for large network planning. The Q-learning method is used to decide the suitable sequence of traffic allocation such that the overall power consumption in the network reduces. In the proposed heuristic, each traffic from the given static traffic demand matrix is successively selected using the Q-learning method and provisioned using the AG-based greedy method.

Negative Effect Prediction and Refueling Traffic Flow Equilibrium of Urban Energy Supply Network during Weekends and Holidays

With the continuous improvement of people’s living standards, the travel demand for vehicles increases rapidly during weekends and holidays. This situation leads to a number of negative impacts on the transport network, such as traffic congestion, carbon emissions, and long queues at refueling stations. The impact of vehicles requiring energy replenishment on the negative effect is nonlinear. Therefore, the negative effect data of working days cannot be directly used to evaluate the cost, and the traffic equilibrium allocation problem also needs to be solved. An equilibrium allocation model of hybrid vehicles considering the energy replenishment demand in holidays is established, aiming at minimizing the total time cost and energy consumption cost. The Frank–Wolfe algorithm is used to solve the traffic allocation problem in advance, and the negative effect of the urban energy network during holidays is predicted by the Genetic Algorithm—Back Propagation (GA-BP) algorithm. Then, the energy-replenishing vehicles in the traffic network are induced to reduce the total cost. Finally, taking the actual road network in Jiulongpo District of Chongqing City, southwest of China, as an example, the negative effects of multiple stations are predicted. The results show that the prediction method proposed in this paper is effective during holiday periods. In addition, as the market share of electric vehicles increases, the negative cost can decrease gradually. The predicted results can provide a reference for traffic managers.

A Hybrid Method of Traffic Congestion Prediction and Control

With the increasing complexity of urban transportation system, serious traffic congestion brings inconvenience to travel. It is also very difficult to predict and control traffic congestion. Therefore, this paper takes urban traffic condition and traffic congestion as the research object, and conducts in-depth research on traffic condition prediction model and traffic congestion control method. Firstly, a traffic state prediction method based on improved particle swarm optimization (IPSO) optimized radial basis function (RBF) and long and short term memory network (LSTM)/ support vector machine (SVM) feature fusion model was proposed for urban traffic state prediction. Experiments were carried out based on the regional traffic data of Shenyang Station, and compared with other algorithms, which verified the superiority of the feature fusion model based on IPSO-RBF and LSTM/SVM in this paper. Secondly, aiming at the problem of urban traffic congestion, a congestion section control method based on traffic allocation is proposed. Through VISSIM simulation and comparison with other traffic control schemes, the superiority of the congestion section optimization method proposed in this paper is verified.

A Collaborative Machine Learning Scheme for Traffic Allocation and Load Balancing for URLLC Service in 5G and Beyond

A Collaborative Machine Learning Scheme for Traffic Allocation and Load Balancing for URLLC Service in 5G and Beyond

Research on the application of search algorithm in computer communication network

Abstract This article mitigates the challenges of previously reported literature by reducing the operating cost and improving the performance of network. A genetic algorithm-based tabu search methodology is proposed to solve the link capacity and traffic allocation (CFA) problem in a computer communication network. An efficient modern super-heuristic search method is used to influence the fixed cost, delay cost, and variable cost of a link on the total operating cost in the computer communication network are discussed. The article analyses a large number of computer simulation results to verify the effectiveness of the tabu search algorithm for CFA problems and also improves the quality of solutions significantly compared with traditional Lagrange relaxation and subgradient optimization algorithms. The experimental results show that with the increase of the weighted coefficient of variable cost, the proportion of variable cost in the total cost increases from 10 to 35%. The growth is relatively slow, and the fixed cost is still the main component. In addition, due to the increase in the variable cost, the tabu search algorithm will also choose the link with large luxury to reduce the variable cost, which makes the fixed cost slightly increase, while the network delay cost and average delay slightly decrease. The proposed method, when compared with the genetic algorithm, has more advantages for large-scale or heavy-load networks.

The Optimal Urban Scale from Different Perspectives and under Different Development Goals Based on the CES Utility Function

To analyze the differences in the optimal urban scale of a country, a government, and residents under different urban development goals and to provide a reference for urban development, an urban development model is constructed based on the constant elasticity of substitution (CES) utility function. The objective function includes four indicators: the economic level, the traffic level, environmental quality, and living conditions. The urban development model is constructed by integrating an urban road planning model, bus route planning model, land use model, and four-stage traffic sharing and traffic allocation model. Based on the setting of different development goals, this model is used to calculate the urban utility value perceived by each stakeholder at different urban scales. Through comparative analysis, we determine the following: (1) When the urban development goals are different, the optimal urban scale of residents and the government differs greatly, and the optimal scale of the country differs very little. (2) When facing the same development goal, the optimal urban scales of the three stakeholders are not the same. However, the three stakeholders are closely related to urban development. Therefore, the selection of the optimal urban scale should comprehensively consider the interests of the three stakeholders.

BGP-Multipath Routing in the Internet

BGP-Multipath (BGP-M) is a multipath routing technique for load balancing. Distinct from other techniques deployed at a router inside an Autonomous System (AS), BGP-M is deployed at a border router that has installed multiple inter-domain border links to a neighbor AS. It uses the equal-cost multi-path (ECMP) function of a border router to share traffic to a destination prefix on different border links. Despite recent research interests in multipath routing, there is little study on BGP-M. Here we provide the first measurement and a comprehensive analysis of BGP-M routing in the Internet. We extracted information on BGP-M from query data collected from Looking Glass (LG) servers. We revealed that BGP-M has already been extensively deployed and used in the Internet. A particular example is Hurricane Electric (AS6939), a Tier-1 network operator, which has implemented >1,000 cases of BGP-M at 69 of its border routers to prefixes in 611 of its neighbor ASes, including many hyper-giant ASes and large content providers, on both IPv4 and IPv6 Internet. We examined the distribution and operation of BGP-M. We also ran traceroute using RIPE Atlas to infer the routing paths, the schemes of traffic allocation, and the delay on border links. This study provided the state-of-the-art knowledge on BGP-M with novel insights into the unique features and the distinct advantages of BGP-M as an effective and readily available technique for load balancing.

Route choice estimation in rail transit systems using smart card data: handling vehicle schedule and walking time uncertainties

Several cities around the world rely on urban rail transit systems composed of interconnected lines, serving massive numbers of passengers on a daily basis. Accessing the location of passengers is essential to ensure the efficient and safe operation and planning of these systems. However, passenger route choices between origin and destination pairs are variable, depending on the subjective perception of travel and waiting times, required transfers, convenience factors, and on-site vehicle arrivals. This work proposes a robust methodology to estimate passenger route choices based only on automated fare collection data, i.e. without privacy-invasive sensors and monitoring devices. Unlike previous approaches, our method does not require precise train timetable information or prior route choice models, and is robust to unforeseen operational events like malfunctions and delays. Train arrival times are inferred from passenger volume spikes at the exit gates, and the likelihood of eligible routes per passenger estimated based on the alignment between vehicle location and the passenger timings of entrance and exit. Applying this approach to automated fare collection data in Lisbon, we find that while in most cases passengers preferred the route with the least transfers, there were a significant number of cases where the shorter distance was preferred. Our findings are valuable for decision support among rail operators in various aspects such as passenger traffic bottleneck resolution, train allocation and scheduling, and placement of services.