Dynamic Scheduling Scheme Research Articles

A GCN-LSTM framework for link prediction in dynamic SIoT networks

The Social Internet of Things (SIoT) paradigm combines the Internet of Things (IoT) with social networking principles, enabling autonomous device interactions. However, the dynamic, heterogeneous, and temporal nature of SIoT networks presents significant challenges for link prediction. While existing models, such as SIoTPredict (Aljubairy et al., 2020), exploits social network analysis and machine learning, they primarily focus only on certain aspects of social relationships between devices, lacking consideration of temporal dynamics, environmental factors, and device heterogeneity. In contrast, we propose a novel framework that integrates Graph Convolutional Networks (GCNs) with Long Short-Term Memory (LSTM) networks and a Multi-Head Attention mechanism, further enhanced by a dynamic feature scheduling strategy, to accurately capture the spatial and temporal dependencies in SIoT environments. Additionally, our work introduces a MATLAB-based SIoT Simulator, providing a robust platform for generating realistic dynamic scenarios and evaluating complex Social Object Relationships. Unlike SIoTPredict, which is optimized for real-time applications but focuses narrowly on social aspects, our approach incorporates a broader range of factors, enabling better predictive accuracy in scenarios with high temporal variability. Extensive co-simulations with MATLAB and PyTorch demonstrate that our framework significantly outperforms existing methods in terms of predictive accuracy, adaptability, and robustness. By bridging critical gaps in current methodologies, this work establishes a comprehensive foundation for future research and practical applications in dynamic SIoT networks.

Dynamic scheduling strategy and algorithm for mixed batch scheduling in vacuum freeze-dried fruit processes

Abstract Vacuum freeze-dried fruit processes consisting of heating and holding are modelled as a mixed batch scheduling with the objective of minimizing the makespan. The jobs differ from each other in job family, size, weight and ready time. The batch processing time is determined by the longest job and the total weight of the jobs in the batch. A mixed-integer linear programming model is developed and tested with small-scale examples. Typical batch scheduling strategies are analysed and a machine-based dynamic programming strategy is proposed. The machine-based dynamic scheduling strategy is applied to design improved genetic and particle swarm optimization algorithms, which demonstrate the effectiveness of this strategy. The worst-case ratio of the algorithms using machine dynamic programming strategy are proved. Numerical experiments show that the heuristic algorithm, genetic algorithm, and particle swarm optimization algorithm based on machine dynamic scheduling strategy outperform related algorithms using greedy and job-based dynamic scheduling strategies.

Application of Distributed Collaborative Optimization in Building Multi-Energy Complementary Energy Systems

This article investigates the application and physical mechanism exploration of distributed collaborative optimization algorithms in building multi-energy complementary energy systems, in response to the difficulties in coordinating various subsystems and insufficient dynamic control strategies. On the basis of modeling each subsystem, the Dual Decomposition algorithm is used to decompose the global optimization problem of the system into several independent sub problems, achieving independent optimization of each subsystem. Through an adaptive dynamic scheduling strategy, real-time data and predictive information are continuously updated and controlled, effectively allocating system resources. The experimental results show that compared to the original system before optimization, the improved algorithm in this paper reduces the total energy consumption of the system by 6.9% and 2.8% on typical summer and winter days, respectively. The conclusion shows that the algorithm proposed in this paper can effectively solve the problem of system coordination difficulties, improve system resource allocation and overall operation level, and provide a new perspective for the optimization design and operation control of energy systems.

Comprehensive Optimal Network Scheduling Strategies for Wireless Control Systems

Although wireless control is one of the key technologies for future industries, most wireless networks are only used for monitoring. When wireless networks are applied to transmit control commands, the uncertain link qualities and limited network resources may destroy the performance of multi-loop control systems. Hence, it is critical to allocate these resources to optimize the control performance as the network condition changes and plants evolve. This article presents comprehensive optimal scheduling strategies for wireless control systems based on adaptive dynamic programming. First, we propose an effective adaptive dynamic programming scheduling (ADPS) strategy to solve the optimal scheduling problem based on the single-step control performance at runtime while significantly reducing computational complexity. Moreover, to overcome the “short-sightedness” of single-step performance prediction, we extend ADPS to ADPS-m ( m ulti-step prediction), which optimizes multi-step performance by incorporating a longer-horizon evolution of the plants. Furthermore, we propose ADPS-H ( H eterogeneous flow scheduling) to support heterogeneous flows with different data rates and sizes and ADPS-H-m ( m ulti-step prediction for H eterogeneous flow scheduling), which schedules heterogeneous flows in a longer prediction horizon. We prove that all these scheduling strategies can achieve optimality and stability under mild assumptions. Extensive experiments integrating TOSSIM and MATLAB/Simulink are performed to evaluate all of the proposed methods in case studies of four- and ten-loop control systems. The simulation results demonstrate that these strategies can effectively improve the control performance at lower computing costs under both cyber and physical disturbances. Under the noise level of \(-\) 76 dBm, for the four-loop case, ADPS achieves the same control performance as the linear programming while saving 99.5% of the execution time. ADPS-m further improves the control performance by up to 27.0% compared with ADPS at the prediction horizon of 3, and ADPS-H-m improves the performance by up to 32.3% and 8.4% compared with round-robin and ADPS-H, respectively. The ten-loop case indicates the effectiveness and scalability of the proposed approaches.

An investigation on energy-saving scheduling algorithm of wireless monitoring sensors in oil and gas pipeline networks

With the rapid development of the oil and gas industry, monitoring the safety and efficiency of pipeline networks has become particularly important. In this context, Wireless Sensor Networks (WSNs) are widely used for monitoring oil and gas pipelines due to their flexible deployment and cost-effectiveness. However, since sensor nodes typically rely on limited battery power, extending the network’s lifecycle and improving energy utilization efficiency have become focal points of research. Therefore, this paper proposes an energy-saving scheduling algorithm based on transformer networks, aimed at optimizing energy consumption and data transmission efficiency of wireless monitoring sensors in oil and gas pipelines. Firstly, this study designs a deep learning-based Transformer model that learns from historical data on energy consumption patterns and environmental variables to predict the energy and data transmission needs of each sensor node. Secondly, based on the prediction results, this algorithm employs a dynamic scheduling strategy that automatically adjusts the sensor’s operational mode and communication frequency according to the node’s energy status and task urgency. Additionally, we have validated the effectiveness of the proposed algorithm through field tests and simulation experiments. According to the experimental results, our model has higher efficiency in energy saving. Compared with Convolutional Neural Networks, Recurrent Neural Networks and Graph Neural Networks, the total energy consumption of sensor networks under the model scheduling in this paper was reduced by 6.7%, 33.4% and 26.3%, respectively. Our algorithms improve the energy efficiency and stability of the monitoring system and provide important technical support for future intelligent pipeline monitoring systems. We hope this paper will inspire future scientific research in this field.

Performance analysis of underwater wireless sensor networks using Reinforcement Learning

This article presents a novel routing protocol named DROR, specifically tailored for underwater wireless sensor networks (UWSNs) to tackle the challenge of void regions. DROR integrates Reinforcement Learning (RL) and Opportunistic Routing (OR) in a recipient-oriented approach, considering the energy limitations and the unique underwater setting. It incorporates a mechanism for void rehabilitation, allowing packets to circumvent void nodes and maintain continuous moving for dependable transmission. Furthermore, a dynamic scheduling strategy based on relative Q-values ensures proficient packet forwarding along the most efficient routing path. Simulation outcomes illustrate the efficacy of the suggested protocol concerning delay, PDR, and energy tax in UWSNs with varying Range, Depths, Packet sizes, and moving radius.

Study on dynamic scheduling method of airport refueling vehicles based on DQN

Aiming at the low utilization rate of airport refueling vehicles and long solution time of exact algorithm caused by the uncertainty of actual flight time, a deep Q network dynamic scheduling method for refueling vehicles combining with the multi-objective deep reinforcement learning framework was proposed. Firstly, an optimization model is established to maximize the on-time rate of refueling tasks and the average proportion of idle vehicles. Then, the five state features that measure the current state of the vehicle are designed as inputs to the network. According to the two objectives, the two scheduling strategies are proposed as the action space so that the algorithm can generate the dynamic scheduling scheme based on the dynamic flight data in real time. Finally, the dynamic scheduling model for airport refueling vehicles is solved, and the effectiveness and real-time performance of the algorithm are verified by different scale examples. The results show that the average number of on-time refueling tasks per day is 9.43 more than that via manual scheduling, and the average working time of vehicles is reduced by 57.6 minutes, which shows the excellent ability of the present method in solving the dynamic scheduling problem of refueling vehicles.

Enhancing V2X QoS: dynamic scheduling scheme over 5G networks and byon

Enhancing V2X QoS: dynamic scheduling scheme over 5G networks and byon

Dynamic Charging Scheduling and Path Planning Scheme for Multiple MC-enabled On-demand Wireless Rechargeable Sensor Networks

With the advancement of wireless energy transfer, Wireless Rechargeable Sensor Networks (WRSNs) have become increasingly popular for efficiently charging sensor nodes. In WRSNs, determining the charging schedule for Mobile Chargers (MCs) is critical for reducing maintenance costs and improving charging efficiency. This is termed the Charging Scheduling Problem (CSP), which is proven to be NP-hard in nature. The existing schemes lack a comprehensive approach to determine the optimal number of MCs and often set a fixed charging threshold for the sensor nodes, degrading charging efficiency in dynamic networks. Additionally, relying on a single MC is unrealistic and impractical for large-scale networks, necessitating a more advanced charging strategy. Thus, this study proposes a dynamic and multi-node charging scheduling scheme named Partitioning-based Charging Schedule for Multiple Mobile Chargers (PCSMMC). The PCSMMC utilizes the traffic load of sensor nodes and energy load of MCs to estimate the optimal number of MCs and computes the progressive threshold for sensor nodes to improve the charging efficiency. Moreover, potential sojourn locations are determined and multiple network factors are integrated into a multi-attribute decision-making process to achieve an efficient charging scheduling and path planning scheme for multiple MCs. The objective of PCSMMC is to enhance the survivability rate of sensor nodes and decrease the traveling path followed by MCs to charge the sensor nodes within the network. Empirical simulation results confirm the superiority of PCSMMC in terms of charging response time, survivability rate, energy utilization efficiency, and network lifetime by a significant margin when compared to alternative approaches.

Dynamic Bayesian game optimization method for multi-energy hub systems with incomplete load information

The interconnection of Energy Hub can enhance the energy efficiency and reliability associated with the independent operation of energy systems. However, the traditional optimal scheduling methods are hard to tackle the situation with the incomplete load information and competitive constraints of a multi-Energy Hub. This paper proposes a dynamic Bayesian game optimization scheduling strategy considering incomplete information on users’ load-side demand. First, a load forecast error coefficient is introduced to handle the conditional probability problem caused by incomplete information through the Bayesian rule. The constraint relationship among multi-Energy Hub under the dynamic price mechanism is analyzed to meet the economic and environmental coordination goals under load uncertainty. Second, extending from Nash equilibrium to Bayes Nash equilibrium proves the existence and uniqueness of Nash equilibrium solutions in Bayesian game models. The decision game algorithm is used to optimize the scheduling of the multi-Energy Hub and subsequently ensure the autonomous scheduling of the system and solve the information barriers. Finally, an Integrated Energy System composed of three Energy Hubs was used to validate the effectiveness and superiority of the proposed optimization method. Results show that under the Bayesian game decision, the multi-Energy Hub game model has reduced the total cost of the Integrated Energy System by 0.78 %, which can improve the economic benefit and address the environmental pollution problem.

Research on Dynamic Scheduling and Route Optimization Strategy of Flex-Route Transit Considering Travel Choice Preference of Passenger

Research on Dynamic Scheduling and Route Optimization Strategy of Flex-Route Transit Considering Travel Choice Preference of Passenger

Dynamic Scheduling of Multi-agent Electromechanical Production Lines based on Iterative Algorithms

In response to the optimization scheduling problem in the dyeing production process, the author proposes a hierarchical scheduling method for dyeing vats based on genetic algorithm and multi-agent. In this method, a hierarchical scheduling algorithm is used to decompose production scheduling into static and dynamic strategies. The static strategy adopts a genetic algorithm that supports batch processing of multiple products, non equality of equipment, order delivery time, switching cost, and other constraints: Dynamic strategy is a coordinated dynamic optimization algorithm that uses multi-agent systems to support the running status of dye tanks based on static strategies. By solving the algorithm with multiple constraints and dynamic factors in the production process, the final result of the dyeing tank operation task is obtained. The simulation compared pure genetic algorithm with manual scheduling, and the results showed that the hierarchical dynamic scheduling strategy based on data-driven achieved the goal of optimizing the production scheduling of dyeing vats. The practical application results also demonstrate the feasibility of this method.

FlowWalker: A Memory-Efficient and High-Performance GPU-Based Dynamic Graph Random Walk Framework

Dynamic graph random walk (DGRW) emerges as a practical tool for capturing structural relations within a graph. Effectively executing DGRW on GPU presents certain challenges. First, existing sampling methods demand a pre-processing buffer, causing substantial space complexity. Moreover, the power-law distribution of graph vertex degrees introduces workload imbalance issues, rendering DGRW embarrassed to parallelize. In this paper, we propose FlowWalker, a GPU-based dynamic graph random walk framework. FlowWalker implements an efficient parallel sampling method to fully exploit the GPU parallelism and reduce space complexity. Moreover, it employs a sampler-centric paradigm alongside a dynamic scheduling strategy to handle the huge amounts of walking queries. FlowWalker stands as a memory-efficient framework that requires no auxiliary data structures in GPU global memory. We examine the performance of FlowWalker extensively on ten datasets, and experiment results show that FlowWalker achieves up to 752.2×, 72.1×, and 16.4× speedup compared with existing CPU, GPU, and FPGA random walk frameworks, respectively. Case study shows that FlowWalker diminishes random walk time from 35% to 3% in a pipeline of ByteDance friend recommendation GNN training.

Stochastic optimization of solar-based distributed energy system: An error-based scenario with a day-ahead and real-time dynamic scheduling approach

Distributed energy systems (DES) have garnered global attention as a promising solution for the expansion of renewable energy sources. However, stochastic uncertainty in renewable sources poses significant challenges in the collaboration optimization of system design and operation. In this study, a comparison analysis was conducted to assess the effectiveness of the conventional distribution-based scenario generation (DS) method for stochastic optimization of a distributed energy system in residential buildings. The results revealed that the DS method inaccurately captured extreme scenarios and exhibited limitations in operation optimization, leading to significant performance evaluation bias. To address these challenges, a novel stochastic optimization approach was developed based on error-based scenarios and a day-ahead and real-time dynamic scheduling strategy (ES-DRS). This approach incorporated solar energy prediction errors to more accurately characterize extreme scenarios, while also considering dynamic dual-scale meteorological boundary condition for rolling operation optimization. Furthermore, the buffer storage and coverage periods in ES-DRS were investigated and discussed during dynamic scheduling. Results demonstrated that when the buffer storage and coverage period were set at 36.22 kWh in summer and 24 h, respectively, the DES with ES-DRS achieved optimal multi-objective performance. This resulted in an annual total cost of 3.21 × 104 USD and CO2 emission of 5.82 tons, representing reductions of 26.45% and 61.06%, respectively, compared to the conventional strategy. Overall, this research contributes to advancing uncertainty analysis and scenario-based optimization in DES, highlighting the potential benefits of adopting the ES-DRS approach to maximize overall performance from both economic and environmental perspectives.

Dynamic Scheduling Scheme With Task Laxity for Data Relay Satellite Networks

Dynamic Scheduling Scheme With Task Laxity for Data Relay Satellite Networks

Derived dynamic scheduling for belief propagation decoding of LDPC codes

AbstractAmong numerous dynamic scheduling strategies for low‐density parity‐check codes, many of them are of high complexity due to repetitive computation and ordering of belief residuals. To address these challenges, two novel approaches are proposed: the derived informed variable‐to‐check residual belief propagation and the derived variable‐node and variable‐to‐check‐edge residual belief propagation for low‐density parity‐check codes. Simulation results show that the derived dynamic scheduling exhibits a substantial gain of approximately 0.15 dB in signal‐to‐noise ratio over the non‐derived one with considerably lower decoding complexity when targeting a bit‐error rate of 10−4.

DAG-Order: An Order-Based Dynamic DAG Scheduling for Real-Time Networks-on-Chip

With the high-performance requirement of safety-critical real-time tasks, the platforms of many-core processors with high parallelism are widely utilized, where network-on-chip (NoC) is generally employed for inter-core communication due to its scalability and high efficiency. Unfortunately, large uncertainties are suffered on NoCs from both the overly parallel architecture and the distributed scheduling strategy (e.g., wormhole flow control), which complicates the response time upper bounds estimation (i.e., either unsafe or pessimistic). For DAG-based real-time parallel tasks, to solve this problem, we propose DAG-Order, an order-based dynamic DAG scheduling approach, which strictly guarantees NoC real-time services. First, rather than build the new analysis to fit the widely used best-effort wormhole NoC, DAG-Order is built upon a kind of advanced low-latency NoC with SLT ( S ingle-cycle L ong-range T raversal) to avoid the unpredictable parallel transmission on the shared source-destination link of wormhole NoCs. Second, DAG-Order is a non-preemptive dynamic scheduling strategy, which jointly considers communication as well as computation workloads, and fits SLT NoC. With such an order-based dynamic scheduling strategy, the provably bound safety is ensured by enforcing certain order constraints among DAG edges/vertices that eliminate the execution-timing anomaly at runtime. Third, the order constraints are further relaxed for higher average-case runtime performance without compromising bound safety. Finally, an effective heuristic algorithm seeking a proper schedule order is developed to tighten the bounds. Experiments on synthetic and realistic benchmarks demonstrate that DAG-Order performs better than the state-of-the-art related scheduling methods.

Research on multi-rider collaborative dynamic scheduling model and algorithm of takeaway delivery

Aiming at the field of high-demand takeaway delivery, a dynamic scheduling strategy based on multi-rider collaborative delivery status is proposed, and the model implementation is obtained by constructing a collaborative dynamic scheduling optimization model and using the optimization TS algorithm to obtain a multi-rider scheduling scheme in a dynamic environment, so that the total delivery time is minimized. Finally, the experimental results show the effectiveness and feasibility of the model and algorithm implementation.

FusionFlow: Accelerating Data Preprocessing for Machine Learning with CPU-GPU Cooperation

Data augmentation enhances the accuracy of DL models by diversifying training samples through a sequence of data transformations. While recent advancements in data augmentation have demonstrated remarkable efficacy, they often rely on computationally expensive and dynamic algorithms. Unfortunately, current system optimizations, primarily designed to leverage CPUs, cannot effectively support these methods due to costs and limited resource availability. To address these issues, we introduce FusionFlow, a system that cooperatively utilizes both CPUs and GPUs to accelerate the data preprocessing stage of DL training that runs the data augmentation algorithm. FusionFlow orchestrates data preprocessing tasks across CPUs and GPUs while minimizing interference with GPU-based model training. In doing so, it effectively mitigates the risk of GPU memory overflow by managing memory allocations of the tasks within the GPU-wide free space. Furthermore, FusionFlow provides a dynamic scheduling strategy for tasks with varying computational demands and reallocates compute resources on the fly to enhance training throughput for both single and multi-GPU DL jobs. Our evaluations show that FusionFlow outperforms existing CPU-based methods by 16--285% in single-machine scenarios and, to achieve similar training speeds, requires 50--60% fewer CPUs compared to utilizing scalable compute resources from external servers.

Partial IDS decoding based on the base graph of protograph LDPC codes

AbstractThe residual belief propagation (RBP) algorithm, which is the most classic informed dynamic scheduling strategy, achieves outstanding performance in error correction and can drastically accelerate convergence speed. However, the greedy algorithmic property of this iterative decoding will inevitably cause loss of decoding performance. To address this, a novel algorithm called the partial average bundle residual belief propagation (PABRBP) is proposed in this paper. According to the construction characteristics of a base matrix of protograph‐LDPC codes, informed dynamic scheduling (IDS) strategies are applied to an edge bundle of base matrices for the first time. This edge bundle of the base matrix can be applied to a corresponding cyclic permutation matrix. Furthermore, the update level of each bundle is determined by the value of the Partially Average Bundle Residual (PABR). Therefore, the edge message with the maximum residual in each bundle is updated in order, and the process of iterative decoding is less likely to become trapped in a local optimum. Additionally, the generation of silent nodes is reduced as much as possible. To further improve the PABRBP decoding performance for medium and long codes over the fading channel, the adjusted compensation term is periodically introduced. Analysis and simulation results show that PABRBP demonstrates a notable convergence quality and decoding performance improvement over the fading channels compared to existing state‐of‐art IDS algorithms.