With the rapid advancement of cloud computing, task scheduling algorithms inspired by natural phenomena have become a research focal point. The grey wolf optimizer (GWO), known for its strong convergence and ease of implementation, has attracted considerable attention. This study introduces an adaptive approach, GWO with the crossover and mutation variant (GWO_C/M), to integrate crossover and mutation strategies and thereby enhance the flexibility and applicability of the GWO. Rather than offering a fixed model, GWO_C/M employs different combinations of crossover and mutation strategies to enhance the balance between exploration and exploitation, solving issues including center bias. Extensive comparisons with 13 state-of-the-art (SOTA) models across six benchmark scenarios showed that GWO_C/M performed robustly, achieving an 87.2% success rate on 41 out of 47 test functions. Moreover, implementing GWO_C/M in CloudSim simulations markedly improved key performance metrics, including total execution time, task completion time, and load balancing. Further validation using the Alibaba Cluster Trace V2018 dataset confirmed that GWO_C/M improved resource utilization and reduced maximum task completion time, indicating the proposed approach's substantial benefits for task scheduling and overall system efficiency in cloud environments. © 2025 The Author(s)

Proactive maintenance is widely recognized for enhancing equipment reliability and reducing downtime costs. However, its role in optimizing spare parts production and distribution decisions remains underexplored, thereby limiting efficient cross-domain resource utilization within the supply-demand system. This paper addresses this gap by studying a maintenance-driven multi-stage joint optimization problem (MMJOP), which integrates flexible spare parts production, multi-vehicle distribution, and imperfect maintenance. We propose an optimal imperfect maintenance strategy to link these cross-domain business activities precisely, and further develop a mathematical model aimed at minimizing energy consumption on the supply side and operational costs on the demand side. To solve the MMJOP, we design an enhanced non-dominated neighbor immune algorithm, featuring a customized initialization operator and a problem-specific local search operator. Additionally, a Qlearning mechanism is employed to automatically select the most appropriate key parameters in the proposed algorithm. Extensive experiments indicate that: (1) the proposed components greatly enhance QNNIA's search performance; and (2) the QNNIA outperforms four well-known comparison algorithms regarding computational optimality, convergence, distribution, and stability. More importantly, the proposed model yields significant economic value, i.e., saving operational costs by 49% with negligible impact on overall energy consumption, proving the necessity of cross-domain business cooperation and resource optimization in the high-end equipment industry.

Requirements analysis is an essential part of the functional planning for cloud platforms. Traditional requirements analysis tends to focus on the customer perspective, rarely considering the dominant role of core manufacturers in the value chain. Integrating service collaboration, sustainable benefits and smart features into the lifecycle service implementation, we first establish the multi-stakeholder, multi-dimensional requirement framework from the perspective of customers, core manufacturers and their partners, including a novel closed-loop business collaboration requirement framework and the sustainable value co-creation framework. Rather than treating value requirements and service requirements as two separate parts, we propose a multi-stage decision model based on Quality Function Deployment (QFD) to establish the connections between service requirements and sustainable value achievement. Thus, the critical service requirements are determined from a holistic perspective of the interaction intensity of the service requirements and their contribution to sustainability. Then, the sensitivity analysis and comparative experiments are performed to verify the effectiveness of the proposed decision model and the necessity of considering judgment reliability in requirements decision. Our research work can guide core enterprises to optimize resource allocation in the value chain, and also provide decision support for the functional module planning of the cloud platform. © 2025 Elsevier Ltd

Efficient maintenance activities are essential for the safe operation of industrial systems, and rational spare parts management, as an integral support to maintenance activities, is also closely linked to operation planning. In this paper, an integrated optimization model of maintenance, spare parts management, and operation for a single-machine multi-component system is proposed, shortened to MSO-SMPS. The goal of MSO-SMPS is the rational design of maintenance strategy, supported by an excellent collaborative management mechanism for new and used spare parts, achieving simultaneous optimization of the total cost and the completion time. Specifically, an adaptive opportunistic maintenance (OM) strategy and a reuse mechanism of retired components are designed to cope with dynamic changes in the system state and operating environment. Combining new and used spare parts can significantly improve the utilization of spare parts while ensuring that maintenance activities are carried out efficiently. In addition, to better address MSO-SMPS, an improved memetic algorithm (IMA) is proposed, in which an initialization method and four local search operators are designed to improve the solve efficiency. Finally, taking the tunnel boring machine (TBM) cutterhead system as a case, extensive experiments verify the effectiveness of the proposed designs. © 2025 Elsevier Ltd

Remanufacturing activities in reverse logistics hold significant theoretical and practical value for their obvious economic and environmental benefits. However, the ever-changing market demand and the uncertain quality of returned items make the management of remanufacturing production highly challenging. Previous studies have mainly focused on determining the optimal remanufacturing decisions based on static market demand and full knowledge of cores quality, while overlooking the dynamic changes in demand information and the imperfect estimation of quality distribution case. Therefore, this paper proposes a two-stage acquisition and remanufacturing method with demand and quality information updating. In the first stage, the remanufacturer formulates acquisition and remanufacturing decisions based on predicted demand and estimated quality distribution. According to updated market demand and the results of the first-stage, the remanufacturer adjusts the optimal decisions for the second stage to maximize profit. In light of this, a two-stage nonlinear mathematical model is established for the acquisition and remanufacturing problem. Based on the scenario analysis method and multivariate optimization theory, optimal strategies for each stage are obtained. To assess the effectiveness of the method put forward, numerical experiments, sensitivity analysis of parameters, and comparative analysis with single-stage acquisition and remanufacturing method are conducted. The results show that the two-stage acquisition and remanufacturing method that accounts for demand and quality information updating can demonstrate greater adaptability to external changes. Compared to the single-stage method, the remanufacturer adopting the two-stage method can achieve an expected profit growth of 6 %∼11 %. Additionally, the effectiveness of the two-stage method is significantly influenced by the reorder point, and there exists an optimal reorder point to maximize the total profit of the remanufacturer. Our research contributes to the uncertainty research in reverse logistics, providing new insights for operational decision-making in remanufacturing enterprises. © 2024 Elsevier Ltd

Electrocardiogram (ECG) prediction is highly important for detecting and storing heart signals and identifying potential health hazards. To improve the duration and accuracy of ECG prediction on the basis of noise filtering, a new algorithm based on variational mode decomposition (VMD) and a convolutional gated recurrent unit (ConvGRU) was proposed, named VMD-ConvGRU. VMD can directly remove noise, such as baseline drift noise, without manual intervention, greatly improving the model usability, and its combination with ConvGRU improves the prediction time and accuracy. The proposed algorithm was compared with three related algorithms (PSR-NN, VMD-NN and TS fuzzy) on MIT-BIH, an internationally recognized arrhythmia database. The experiments showed that the VMD-ConvGRU algorithm not only achieves better prediction accuracy than that of the other three algorithms but also has a considerable advantage in terms of prediction time. In addition, prediction experiments on both the MIT-BIH and European ST-T databases have shown that the VMD-ConvGRU algorithm has better generalizability than the other methods. © 2024, The Author(s).

Typically, user behaviour occurs continuously, and considering this dynamic sequence correlation can lead to more accurate recommendations. Sequential recommendation systems have, therefore, become an important means of solving the problem of network information overload. However, existing attention mechanisms are still insufficient for modelling users’ dynamic and diverse preferences. This paper presents a recommendation model based on a multiheaded self-attention mechanism and multitemporal embeddings of long- and short-term interests (MSMT-LSI). MSMT-LSI balances users’ long- and short-term benefits through two multihead self-attention networks and finally forms a hybrid representation for recommendation. After finding the most suitable parameter combinations for the MSMT-LSI model through parameter sensitivity analysis and verifying the advantages of the long- and short-term fusion strategy, related experiments on five well-known datasets and their analysis shows that the performance of MSMT-LSI is better than that of the classical model on the same benchmark dataset. © The Author(s) 2024.

Outsourcing machine maintenance to third parties has been a trend due to the increasing complex of machines and the benefits of maintenance outsourcing. However, this new phenomenon is ignored in previous studies pertaining to the integrated optimization of production and maintenance scheduling (IOPMS) problems, resulting the lack of theoretical guidance for production managers to formulate optimal scheduling schemes under this new trend. In this study, we investigate an IOPMS problem considering maintenance outsourcing in a distributed parallel machine environment, referred as IOPMSTW, which requires the use of third-party worker resources to perform preventive maintenance. Makespan and total cost are two optimization objectives. We first formulate the problem by developing a mixed integer linear programming. Then a memetic algorithm incorporating iterated greedy method (IG) is proposed to solve the IOPMSTW, in which an improved decoding method and a problem-dependent local search operator based on IG are designed to respectively ensure the feasibility of new generated individuals and improve the searching efficiency. The validity of proposed mathematical model is verified by CPLEX based on eight small instances. Based on 240 constructed instances, a set of comprehensive experiments are conducted. The results demonstrate that the local search operator improved the searching performance of the proposed algorithm by 100%. Comparison results with other well-known algorithms show that the proposed algorithm achieved the best results on more than 85% of the tested instances. © 2024 Elsevier Ltd

The existing research on integrated production scheduling typically focuses on activities related to both production and post-production (e.g., operation and maintenance), with limited consideration for simultaneously integrating pre-production activities (e.g., raw material procurement). However, achieving the equilibrium between the manufacturer and the demander for intelligent manufacturing systems requires the optimal scheduling solution that integrates both pre-production and post-production activities. Inspired by this, we investigate a novel integrated scheduling problem that concurrently considers raw material procurement, production scheduling and equipment maintenance (abbreviated as SIPPM). A mixed-integer linear programming model is developed to simultaneously minimize the total costs for the manufacturer and the demander. Furthermore, a double-layer Q-learning driven memetic algorithm (DQMA) is proposed to solve the SIPPM. In DQMA, a welltailored three-layer hybrid encoding method is presented for chromosome representation. The global search of DQMA employs three crossover and three mutation operators. Moreover, a knowledge-based local search operator with six methods, guided by an effective double-layer Q-learning structure, is devised to enhance local exploitation capabilities. The superiority of DQMA is verified through comparison with three popular multiobjective optimization algorithms on 108 newly established benchmark instances. The proposed integrated scheduling mode is proven to be more effective than two separated scheduling modes without considering raw material procurement.

Dynamic events and transportation constraints would significantly affect the full utilization of resources and the reduction of production costs in distributed job shops. Therefore, in this paper, a deep reinforcement learning algorithm (DRL)-based real-time scheduling method is developed to minimize the mean tardiness of the dynamic distributed job shop scheduling problem with transfers (DDJSPT) considering random job arrivals. Firstly, the proposed DDJSPT is modeled as a Markov decision process (MDP). Then, ten problem-oriented state features covering four aspects of factories, machines, jobs, and operations are elaborately extracted from the dynamic distributed job shop. After that, eleven composite rules considering the uniqueness of DDJSPT are constructed as a pool of actions to intelligently prioritize unfinished jobs and allocate the selected job to an appropriate factory. Moreover, a justified reward function adapted from the objective is designed for better convergence of DRLs. Subsequently, five DRLs are employed to address the DDJSPT, encompassing deep Q-network (DQN), double DQN (DDQN), dueling DQN (DlDQN), trust region policy optimization (TRPO), and proximal policy optimization (PPO). Finally, grounded in numerical comparison experiments under 243 production configurations of the DDJSPT, the effectiveness and generalization of DRL-based scheduling methods are credibly verified and confirmed. © 2024 Elsevier Ltd