Search Results (634)

Limited by the computational performance limits of the CPU(Central Processing Unit), the traditional Spark architecture struggles to achieve high throughput and low latency under the dual pressure of a large data scale and real-time requirements in centralized control systems. This work uses a publicly available CNC(Computer Numerical Control) milling dataset as a functional validation proxy for time-series data processing, then extends validation to a large-scale synthetic power transmission grid dataset. Furthermore, Spark-GPU(Graphics Processing Unit) collaboration suffers from load balancing failure due to heterogeneous resource scheduling and communication overhead, thus failing to unleash its performance potential. This paper proposes a Spark-GPU fusion acceleration technology path. The path consists of three key components: first, it integrates the RAPIDS accelerator; second, it designs a GPU-aware partitioning and task co-scheduling strategy; and third, it optimizes the zero-copy data path. Together, these components realize an integrated collaboration of heterogeneous resources. Validation on real-world datasets yields the following results. In real-time aggregation scenarios, the proposed solution improves throughput by a factor of 3.7 over the pure CPU baseline and reduces end-to-end latency by 62%. Compared with the basic GPU solution, GPU utilization rises from 51.7% to 72.3%, representing a relative improvement of 39.8%. Furthermore, the solution meets industrial-grade high availability requirements. This research significantly improves the processing throughput and reduces end-to-end latency in typical centralized control scenarios, thus providing a feasible technical route for demanding concurrent centralized control scenarios such as electric power industry manufacturing with high real-time demands. Full article

In the fabric warehouse, order picking operations face high labor intensity and rising operational costs, requiring urgent optimization. This study investigates the order scheduling and task assignment problem within an elastic staffing framework, where temporary labor recruitment and real-time task allocation need to be adjusted dynamically in response to fluctuations in order volumes. Nevertheless, conventional approaches often suffer from severe computational bottlenecks under such highly dynamic conditions, and struggle to maintain optimal solutions when demand undergoes large and frequent fluctuations. To address these challenges, this study proposes a Graph Transformer Policy Network with Proximal Policy Optimization (GTP-PPO), which combines graph structure features with a global attention mechanism. First, the return picking strategy and the S-shaped picking strategy are compared and analyzed in the fabric warehouse scenario. The results reveal that the return strategy is more suitable for the studied warehouse layout. Subsequently, a mixed-integer programming (MIP) model and a GTP-PPO model are established for optimizing order dispatching and scheduling. Finally, an empirical analysis is carried out based on the peak order day of the year in the fabric warehouse. The results demonstrate that the proposed GTP-PPO model not only achieves near-global optimal solutions (gap < 4%) comparable to the MIP model, but also exhibits robust real-time decision-making capabilities under dynamically increasing order volumes and unexpected disruptions. Compared to the MIP model, the GTP-PPO approach reduces unskilled labor hours by 84.80% and decreases operational volatility by 27.60%, with only a 3.52% increase in operational costs. Full article

Data centers have become essential infrastructure for digital services, while their rapidly growing electricity demand makes coordinated workload and power management an important optimization problem. This paper studies the multi-data-center operation problem under time-of-use electricity pricing and formulates it as a multi-data-center mixed-integer nonlinear programming model (MDC-MINLP). The model jointly represents binary task scheduling decisions, including temporal workload shifting and spatial task migration, and continuous power-side variables, including device-level utilization, IT and auxiliary power consumption, energy storage dynamics, grid power procurement, and quality-of-service constraints. The objective is to minimize the total operating cost by integrating electricity purchasing cost, IT operation loss, storage degradation cost, and migration cost. To solve the resulting large-scale discrete–continuous coupled problem, an Adaptive Large Neighborhood Search algorithm with Hierarchical Collaboration (HC-ALNS) is proposed. HC-ALNS reconstructs feasible task action sets, employs a surrogate objective for fast candidate screening, performs accurate power-layer evaluation for selected solutions, and adaptively adjusts search intensity according to convergence behavior. Numerical results show that HC-ALNS reduces the total operating cost by 3.67% and achieves better convergence and solution quality than NSGA-II and PSO. These findings demonstrate that the proposed MDC-MINLP and HC-ALNS provide an effective mathematical optimization framework for coordinated computation–power scheduling. Full article

Spaceborne neural processing units (NPUs) increasingly support real-time deep-learning inference, but their dense multiply-accumulate arrays are vulnerable to radiation-induced soft errors. Conventional radiation-hardening methods improve reliability through hardware redundancy, but they incur substantial area, performance and compiler-mapping overheads. This paper proposes tile-resilient algorithm-based fault tolerance (TR-ABFT), a software-scheduled, detection-oriented scheme for quantized NPU inference. TR-ABFT generates checksum information at tile granularity and maps checking tasks onto the original processing element (PE) array without changing the hardware topology. To make ABFT compatible with INT8 datapaths, we design two checksum-coding strategies: checksum decomposition and modulo-239 checksum coding. The modulo-239 scheme removes structural missed detections for two-bit flips with bit-position spacings in (1, 31), while preserving compatibility with signed INT8 inputs. Evaluations on ResNet, YOLOv8, and RT-DETR show that, on a $16\times 16$ array, TR-ABFT introduces only 6.37% to 24.61% additional computational overhead. By converting spatial redundancy into schedulable temporal redundancy, TR-ABFT preserves systolic-array regularity and provides a low-overhead reliability-enhancement mechanism for space-grade neural-network accelerators. Full article

With the rapid expansion of artificial intelligence infrastructure and cloud computing services, data centres are evolving from rigid electricity loads into flexible resources capable of contributing to renewable energy integration, grid regulation and cross-regional computing power allocation. Addressing the shortcomings in existing research regarding the differences between various types of computing tasks, the mechanisms of green migration under network constraints, and the characterisation of curtailment risks for renewable energy, this paper proposes a risk-aware collaborative optimisation and scheduling method for a power–computing–network–storage coupled system across remote data centres. Firstly, a hierarchical model of multi-type computing tasks is constructed, classifying data centre loads into fixed real-time tasks, online inference tasks, long-duration AI training tasks, and opportunistic elastic tasks, to characterise the differences between these tasks in terms of latency, time-shift, migration, and completion volume constraints. Secondly, a graph-attention-inspired green affinity prior is proposed, mapping grid topological distance, renewable energy availability, data centre PUE, and energy storage regulation capacity into interpretable migration signals, thereby guiding flexible computing power to migrate towards nodes with abundant green electricity and favourable grid support conditions. Subsequently, we introduce the CVaR metric to quantify the tail risk of renewable energy curtailment, establishing a multi-scenario stochastic linear optimisation model that incorporates DC power flow, unit output, renewable energy utilisation, campus energy storage, task SLAs, and cross-node migration constraints. A 24 h simulation based on the IEEE 10-machine, 39-node system demonstrates that the proposed method can reduce the expected curtailment volume from 176.939 MWh to 0 MWh, lower the CVaR curtailment risk from 694.085 MWh to 0 MWh, and increase the proportion of green computing power by 9.283 percentage points compared to the fixed-load baseline, whilst improving the five-tier collaborative score by 4.885 points. Full article

Driven by global population aging, long-term in-home and institutional elderly care faces challenges in delivering continuous, privacy-aware, and resource-efficient safety and health monitoring. Existing edge-based solutions struggle to jointly balance detection accuracy, privacy, and resource overhead during continuous operation, and often have limited situational awareness and inflexible management. We propose EdgeElderCare, a resource-aware, scene-adaptive edge-cloud collaborative system for continuous elderly safety and health monitoring. Its contributions are threefold: (1) a scene-adaptive multi-sensor task-sharing architecture that deploys vision-based fall detection in public areas and privacy-aware millimeter-wave radar in private spaces. Combined with edge-side task scheduling, it provides spatially complementary coverage of public and private areas, mitigates the accuracy–privacy conflict, and reduces computing and bandwidth consumption relative to data-level fusion; (2) a lightweight myocardial infarction detection module deployed on an edge platform, enabling local ECG analysis with low resource overhead; (3) a 3D digital-twin edge-cloud management platform that maps multi-source sensing data to a virtual scene in real time and supports hierarchical visual alerting. Experiments in a real nursing home environment show that the system operated stably on resource-constrained edge hardware: UWB positioning achieved centimeter-level RMSE, visual fall detection reached a recall of 0.90, millimeter-wave radar fall detection achieved accuracy, and F1 above 0.90, and myocardial infarction detection exceeded 0.99 accuracy on the public PTB/PTB-XL benchmark. These results indicate an engineering-feasible approach to intelligent elderly care. Larger-scale and longer-term validation remains the focus of future work. Full article

Mobile edge computing (MEC) enables computation-intensive and latency-sensitive tasks to be offloaded from mobile devices to nearby edge servers. Most existing MEC task offloading studies formulate offloading as a selection problem over a fixed or fully available set of candidate servers, which is restrictive in heterogeneous MEC scenarios with task-node eligibility constraints. Under such constraints, a task can be processed by an edge server only when task attributes, service requirements, link conditions, and node states jointly satisfy the corresponding eligibility conditions. The feasible action set therefore varies over time, while offloading decisions are further coupled with edge-node-side queue competition and long-term load evolution. To address this problem, this paper proposes Resource-oriented Scheduling Coordination (RoSCo), a load-aware task offloading method with scheduling-level constraint handling for eligibility-constrained MEC systems. In this paper, scheduling coordination refers to the joint use of feasible-action control, priority-aware edge-node service-order modeling, and load-responsive feedback within the task offloading decision process; it does not denote inter-server communication, task aggregation, federated model aggregation, or a distributed coordination protocol. RoSCo constructs a dynamic feasible action set, applies eligibility-aware action masking to exclude infeasible offloading actions, incorporates priority-aware edge-node service-order information to characterize queueing competition among heterogeneous tasks, and designs a load-responsive reward to guide congestion mitigation and load balancing. A dueling double deep Q-network (D3QN) is adopted as the value-learning backbone, while the main methodological contribution lies in embedding task-specific feasible-action control, priority-aware node-side queue information, and load-responsive feedback into the constrained offloading process. Simulation results show that RoSCo reduces the task drop rate and edge-node load imbalance while maintaining competitive task completion delay and energy consumption, especially under high-load and sparse-eligibility conditions. Full article

As scientific workflows grow more complex and green computing becomes a priority, efficient multi-objective scheduling is essential to optimize makespan, cost, and energy consumption for large task graphs. However, existing methods often suffer from scalability bottlenecks and insufficient modeling of task dependencies, leading to degraded performance on large-scale workflows. This paper proposes HiGAT-AC, a framework that combines a hierarchical graph attention network with actor-critic reinforcement learning for scalable workflow scheduling in heterogeneous systems. HiGAT-AC splits large workflows into subgraphs via spectral clustering and uses a three-level hierarchy to capture local task dependencies, coordinate inter-subgraph information, and conduct global resource allocation. The actor-critic model employs Chebyshev scalarization to balance the three conflicting objectives. Experimental results show that HiGAT-AC achieves competitive composite scores across workflow scales from 500 to 1000 tasks, with scores reaching 0.954 on 500-task workflows and 1.000 on 1000-task workflows, while remaining stable above 0.70 across all scales. Compared with traditional and representative learning-based methods, HiGAT-AC exhibits favorable overall performance and relatively stable scalability on large task graphs, providing a promising solution for scientific workflow scheduling that balances performance and sustainability. Full article

Textile equipment monitoring and scheduling are constrained by device heterogeneity, stringent real-time requirements, and complex dynamic resource scheduling. To address these challenges, this study proposes a cloud-edge-end collaborative remote monitoring and scheduling system for textile equipment. The proposed system aims to overcome the limitations of traditional solutions in compatibility, real-time performance, and resource utilization. This work is positioned as an applied systems study, in which the scheduling modules are used as monitoring-driven service extensions rather than as standalone algorithmic contributions. We develop (i) an adaptive multi-protocol parsing mechanism, (ii) a collaborative hierarchical alerting framework, and (iii) monitoring-driven computing-resource and production-scheduling services. The system is implemented across the terminal device layer, edge computing layer, and central cloud layer. Embedded acquisition terminals were designed to support multiple industrial protocols, including Modbus RTU, OPC UA, and EtherCAT. Dynamic protocol adaptation was used to identify, parse, and map heterogeneous protocol frames into a unified information model at runtime. In the workshop deployment reported in this study, field validation was conducted on 120 air-jet looms connected through RS485-based Modbus RTU. Other interfaces were evaluated as prototype-supported communication options rather than as quantitatively validated workshop interfaces. A cloud-edge-end collaborative alerting framework is designed by combining an improved OPTICS algorithm with a graph neural network (GNN) model. It improves the redundant-alarm filtering rate by 42.1%, achieves 96.8% root-cause diagnosis accuracy, and keeps the end-to-end alert latency at or below 200 ms at the 99th percentile. A cross-layer resource scheduling strategy incorporating a fuzzy PID controller is proposed, accompanied by a weighted multi-criteria resource-optimization model. This strategy increases the average CPU utilization of edge nodes to 84.3 ± 3.6% and reduces burst-task response latency to 236 ± 48 ms. In addition, an adaptive particle-swarm optimization module based on a scalarized composite scheduling objective reduces the equipment idle rate to 6.5% and shortens the average order completion time by 28.4%. Overall, the proposed framework demonstrates the feasibility of cloud-edge-end collaborative monitoring and scheduling in the validated RS485/Modbus-RTU-based weaving-workshop scenario, while its application to other textile processes, machine types, and communication configurations requires further protocol-specific adaptation and field validation. Full article

The Exponential Distribution Optimizer (EDO) is a newly developed mathematics-based metaheuristic with a simple structure and high efficiency. However, the EDO faces dilemmas, including poor initial population quality, premature convergence, insufficient population diversity, and low convergence accuracy when addressing complex high-dimensional optimization and cloud computing task scheduling problems. To overcome these drawbacks, this paper proposes an Adaptive Synergistic Exponential Distribution Optimizer (ASEDO) integrated with three collaborative strategies for global optimization and cloud computing task scheduling. First, a Multi-Source Hybrid Perturbation Initialization is designed using first-order differential mutation and high-order Bernstein polynomial perturbation to expand the initial search space and boost population diversity. Second, a Bipolar Adaptive Search Mechanism is presented to enable bidirectional learning from elite and inferior individuals, effectively preventing local optima trapping. Third, an Oscillating Random Mapping Learning Mechanism is introduced to strengthen local search ability and convergence precision via random learning and second-order oscillation mapping. The proposed ASEDO is verified on CEC2022 benchmark functions and cloud computing task scheduling under small-scale, large-scale, and dynamic task scenarios. Ablation experiments and comparison results demonstrate that the synergistic effect of the three strategies significantly improves the performance of EDO. Meanwhile, the ASEDO shows stronger global search capability, higher solution accuracy, and better stability than several state-of-the-art algorithms in both global optimization and cloud task scheduling applications. Full article

Deploying Large Language Models (LLMs) in collaborative multi-agent settings represents a promising frontier for complex AI problem-solving, yet the field lacks systematic mechanisms to manage the inherent coordination overhead and resource contention that arise at scale. Existing LLM-based Multi-Agent System (MAS) frameworks predominantly adopt sequential or loosely coupled execution models, which fail to exploit the parallelism potential of modern computing environments and limit overall system throughput. To bridge this gap, this paper presents DynTaskMAS, a framework that redefines task orchestration in LLM-based MASs through a dynamic task graph abstraction. Rather than treating tasks as static pipelines, DynTaskMAS continuously models task interdependencies at runtime, enabling opportunistic parallel execution while preserving logical correctness. The architecture integrates four synergistic components: a runtime task decomposition module that captures evolving dependencies among subtasks; a scheduling engine that dispatches ready tasks to available agents without centralized bottlenecks; a context propagation layer that maintains shared semantic state across concurrently executing agents; and a self-tuning workflow controller that adapts execution priorities based on observed system load. Together, these components address a core tension in LLM-based MAS design, balancing agent autonomy with coordinated efficiency. Evaluations across tasks of varying complexity confirm that DynTaskMAS delivers substantial gains in execution efficiency (21.3–33.0% reduction), resource utilization (from 65% to 88%), and agent scalability (3.47× throughput with 16 concurrent agents) compared to sequential baselines. This work offers a generalizable architectural blueprint for next-generation LLM-based Multi-Agent Systems operating under real-world dynamic and resource-constrained conditions. Full article

In edge-intelligent systems, efficient resource management and task scheduling are critical but challenging due to the dynamic and heterogeneous nature of edge nodes (e.g., IoT devices, drones). We model this dynamic resource allocation challenge as an online sleeping Restless Multi-Armed Bandits (RMAB) problem, where each edge node (arm) operates as a Markov decision process. Unlike prior RMAB frameworks assuming perpetual availability, our setting captures the stochastic availability of edge nodes across rounds. The system controller (learner) is unaware of the transition functions, reward distributions, and node availability a priori. The goal is to maximize expected cumulative rewards through adaptive node selection. To explore this target problem, we first derive an asymptotically optimal sleeping-index policy (SIP) as the oracle based on the fluid process transformation. Then we propose OSILA (Online Sleeping Index-aware Learning Algorithm), featuring a Minimum Exploration Guarantee (MEG) mechanism for efficient exploration. This is coupled with a modified Linear Programming-based exploitation mechanism to construct an online sleeping index, effectively handling dynamic node availability. To the best of our knowledge, this work is the first to provide the theoretical analysis (which achieves $\tilde{\mathcal{O}}\left(K{T}^{2/3}\sqrt{logT}\right)$ regret where K is the number of arms and T is the time horizon) to the online sleeping RMAB problem. Empirical results validate both theoretical guarantees and practical effectiveness in dynamic edge computing environments. Full article

While artificial intelligence (AI) can improve energy efficiency in carbon neutrality applications, its high energy consumption and rebound effect weaken the actual emission reduction effect. To address the issues of high energy consumption and the rebound effect of AI weakening emission reduction, this paper proposes a green AI-driven environmental economic computing framework. First, an energy consumption perception index is introduced, and carbon emissions are monitored in real time using Carbontracker version 2.4.2. Second, multi-task learning is used to predict energy demand and emissions based on multi-source data. Third, the rebound effect is quantified and corrected using an elasticity coefficient model. Finally, resource allocation is optimized under environmental constraints through reinforcement learning, and a closed-loop feedback mechanism is constructed. Experimental results show that the carbon emissions from GPT-3 training are as high as 590 kgCO₂, while the emissions from YOLOv5 are only 59 kgCO₂. Dynamic batch processing improves energy efficiency by 45%, and the knowledge distillation rebound index is 0.75, but the net energy-saving rate is only 9.1%. The information technology industry achieved a synergy index of 0.88 through AI optimization, but the response time of dedicated hardware is 1.5 s (three times faster than the cloud), indicating that large-scale models have high energy consumption and need optimization to prevent rebound. Real-time feedback and hardware scheduling are key to achieving carbon neutrality. Full article

Artificial Intelligence (AI) is revolutionizing Construction Management (CM) through automation, predictive analytics, and real-time decision-making throughout the project lifecycle.This study aims to provide a comprehensive and structured synthesis of AI models and their applications in CM. This paper presents a systematic review of 191 peer-reviewed articles published between 2020 and 2025, aiming to integrate the current state of AI implementation in CM, focusing on AI methods and models and their applications in CM. Compared to previous reviews that take these factors individually or focus narrowly on specific techniques, this study offers a comprehensive taxonomy that systematically maps AI techniques against CM functions and integration platforms. The results reveal that AI applications are primarily concentrated in risk and safety management, decision support, and monitoring and control, while domains such as legal analytics, robotics, and cybersecurity remain underexplored. Furthermore, Computer Vision (CV) and Deep Learning (DL) dominate tasks such as safety monitoring and defect detection, whereas Machine Learning (ML) and optimization algorithms are widely applied in cost estimation and scheduling. It also addresses developments rarely covered in construction research, including Generative AI (Gen-AI), Explainable AI (XAI), and transformer models, presenting a strategic framework for the widespread adoption of AI in the construction environment. This study contributes a structured taxonomy that systematically links AI models with CM functions and enabling technologies, providing a comprehensive synthesis of emerging trends and research gaps. Full article

Aircraft maintenance scheduling is a critical task in air transportation and national defense security, characterized by complex multi-step procedures, strict precedence dependencies, and multi-resource constraints involving personnel skills and equipment availability. Traditional scheduling methods and standard metaheuristic algorithms often suffer from insufficient model adaptability, poor population diversity, premature convergence, and complex encoding schemes that require frequent feasibility checks. To address these challenges, this paper proposes a comprehensive optimization framework based on the Resource-Constrained Project Scheduling Problem (RCPSP) model. A decimal priority-based encoding method is introduced to replace traditional integer permutation encoding, significantly reducing computational complexity and enhancing search space continuity. Furthermore, an improved hybrid Particle Swarm Optimization algorithm integrating reverse learning and partial random operations (RL-PSO) is developed. The reverse learning mechanism expands the global search space by generating reverse particles, while partial random operations maintain population diversity and prevent premature convergence. The proposed framework converts priority encoding into feasible schedules through a priority sorting and left-shift resource allocation strategy. Simulation experiments on maintenance tasks involving up to 50 aircraft demonstrate that RL-PSO achieves optimization accuracy of 332 min, convergence speed of 92.07 s, and stability of 2.8843 min in standard deviation, which are superior compared to standard PSO, Simulated Annealing, and Teaching–Learning-Based Optimization combined with the serial schedule generation scheme (SSGS). The method effectively balances global exploration and local exploitation, making it suitable for complex, large-scale aircraft maintenance scenarios. Future work will extend the framework to multi-objective optimization and dynamic scheduling environments. Full article