Product design involves the conceptualization and creation of products, where simulation workflows often require dynamic scheduling. Although deep reinforcement learning has shown competitive performance in addressing dynamic scheduling problems, they encounter challenges when dealing with multiple types of dynamic events in workflow scheduling. To this end, this article introduces a novel meta-reinforcement learning-based dynamic scheduling algorithm (MDSA) to tackle multiple dynamic events. Four types of dynamic events are considered and modeled, including random insertion, loop execution, structural adjustments, and resource uncertainties. The Markov decision process is formulated for the workflow scheduling problem, defining the state, action, reward, and state transition. To capture time-sequential information, a context-aware discretized policy is introduced, and a meta-training algorithm is employed to extract common knowledge across various scheduling scenarios. The effectiveness of the proposed method is demonstrated through a case study in the semiconductor display industry. The simulation environments are constructed using industry-inspired synthetic data, with reference to deployment practices from BOE and task patterns derived from Alibaba cluster traces, to ensure practical relevance while maintaining experimental control. Experimental results indicate that the approach demonstrates strong generalization capabilities, especially in terms of its consistent performance across varying task data distribution and its robustness in adapting to multiple dynamic conditions, thereby enhancing its adaptability in practical applications.

The Industrial Internet of Things provides an opportunity for flexible and collaborative manufacturing, but introduces more risk and more communication overhead from the Internet to the industrial field. To avoid attacks from unreliable service providers and requesters, Industrial Demilitarized Zone (IDMZ) is introduced in conjunction with firewalls to provide new communication modes between edge servers and industrial devices. As the number of tasks being offloaded to the edge side increases, optimal task offloading to balance the risk and the communication overhead with limited demilitarized buffer size becomes a challenge. Therefore, this paper establishes a mathematical model for secure task offloading in the Industrial Internet-of-Things considering dense communication with different communication modes. Then, a Parallel Gbest-centric differential evolution (P-G-DE) is designed to solve this task offloading problem with a heuristic-embedded initialization strategy, a modified Gbest-centric differential evolutionary operator and a circular-rotated parallelization scheme. The experimental results verify that the proposed method is capable of providing a high-quality solution with a lower risk and a shorter execution time in seconds, compared to six state-of-the-art evolutionary algorithms.

Multi-stage simulation workflows are a commonly used and crucial means for evaluating and optimizing design performance throughout the stages of product development. Effective multi-workflow scheduling is crucial for ensuring optimal resource utilization and execution time. Addressing simulation multi-workflow dynamic scheduling (SWDS) is challenging due to the variability of tasks and the uncertainty in execution, necessitating flexible and adaptive scheduling strategies. While traditional methods such as heuristic-based algorithms are popular in workflow scheduling, they show weaknesses in robustness, generalization, and adaptability when dealing with highly dynamic environments such as SWDS. To address above issues in SWDS, this paper innovatively proposes a meta-reinforcement learning-based scheduling method that aims to enhance generalization and adaptability to dynamic conditions. An enhanced Model-Agnostic Meta-Learning based deep reinforcement learning (DRL) algorithm is proposed to acquire dynamic scheduling strategies through multi-scenario training. Multi-step state features are extracted to address the issue of insufficient state observations. Conjugate adaptive search and Armijo conditions are applied to enhance the effectiveness of algorithm training. Experimental tests in 180 multi-type scenarios, compared with nine heuristic methods and three state-of-the-art DRL algorithms, comprehensively demonstrate the superiority of the proposed method.

Artificial intelligence-generated content (AIGC) is an automated method that generates the content according to its knowledge and the intent information. Deploying the AIGC model in edge networks unlocks new possibilities. Unfortunately, realizing AIGC in distributed edge networks faces critical challenges, including the interdependent control decisions, the complex trade-offs between energy consumption and total variation (TV) distance, the extension of TV distance due to insufficient denoising steps, the restriction of AIGC inferencing deadlines, and the uncertainty of AIGC task arrivals. In this paper, targeting AIGC tasks, we design polynomial- time online algorithms to overcome all these challenges. Firstly, we formulate distributed AIGC as a non-linear mixed-integer program for long-term total cost optimization. Subsequently, we propose a novel algorithmic approach that generates candidate inferencing schedules, reformulates the original problem into anew schedule selection problem, and solves this new problem using an online primal-dual-based algorithm. Moreover, we rigorously prove that our approach leads to a constant competitive ratio for the long-term total cost. Through extensive evaluations using real-world data, the superior practical performance of our approach is demonstrated, reducing the total cost by more than 50% compared to various alternative methods.

Agent-based simulation has become essential for simulating complex evolving systems, such as social systems and complex manufacturing systems. However, due to the uncertainty in agent behaviors, model calibration poses a significant challenge. Traditional methods like try-and-error and regression are inefficient, requiring numerous simulation runs under different parameter settings. To address this, we propose an online pattern clustering network-based calibration method, termed PCN-Calibration. This method establishes a pattern clustering network using reference data from the target system and compares simulation results under random parameter settings with this network. Ten weighting rules are introduced to estimate the best parameters based on the distance between the simulation results and the clustering network. Experimental results demonstrate that the proposed method can find feasible parameter settings within seconds. Furthermore, the experimental discussions provide guidelines for selecting suitable weighting rules for calibrating different models.

Online simulation task scheduling in a private cloud manufacturing platform usually requires rapid decision-making algorithms because of the characteristics of unpredictability and diversity of tasks. However, the existing approaches face challenges in generating satisfactory scheduling schemes within a limited solving time. Therefore, this paper proposes a dynamic scheduling algorithm for online simulation task scheduling that is based on cross-attention and deep reinforcement learning (DRL). A multichannel DRL-based framework with discrete event triggering is introduced to effectively recognize online scheduling environments. An innovative multistep state feature cross-attention method is proposed to address the challenge of temporal features caused by nonsimultaneous task arrivals. A case study in the semiconductor display industry with 35 diverse scheduling scenarios was conducted to evaluate the efficacy of the proposed algorithm, which was compared with six classic state-of-the-art DRL algorithms and three commonly used priority dispatching rules. The results show that the proposed algorithm maintains superior scheduling performance across multiple scheduling scenarios and outperforms the other algorithms by an average of nearly 30

Service composition enables the flexible and agile collaboration of multiple services to complete personalized manufacturing tasks in cloud manufacturing. Compared with traditional manufacturing mode and cloud computing, trust problems become more serious and crucial in cloud manufacturing because of the nontransparency and short-term cooperation mode. A trust evaluation method for service composition in cloud manufacturing is proposed in this article. To quantitatively calculate the trust, a trust evaluation index system is established that comprehensively considers the influencing factors in the production, transaction, and collaboration processes of cloud manufacturing services. The trust value is synthesized based on all index values using the criteria importance through intercriteria correlation method. To extract the temporal information in historical trust data, a time-aware predictive trust evaluation method based on gated recurrent unit is proposed to learn the changing pattern of trust value over time. The trust data are trained together with their timestamps to predict the trust in the scheduled transaction time. The correlation between services in service composition is modeled to mine the correlation information by association analysis. The trust of service composition depends on the trust values of all component services and the correlations between them. The experiments demonstrate the effectiveness of the proposed method through case studies and performance comparisons with other state-of-art methods.

Distributed edge computing systems that participate in emergency demand response (EDR) programs can adjust workload across heterogenous edges to reduce total energy consumption. Unfortunately, this approach may not always reduce sufficient energy as required by EDR. In this article, we propose to leverage Electrical Vehicles (EVs) and Vehicle-to-Grid (V2G) techniques to provide energy to the edge system, and design an auction mechanism to incentivize EVs to discharge energy for the edges. Yet, we face critical challenges, such as the uncertainty of EV bid arrivals, the restriction of discharging deadlines, and the desire to achieve required economic efficiency. To overcome such challenges, we design a novel online approach, (EDR)-D-3, of multiple algorithms that decompose our original NP-hard social cost minimization problem into two subproblems, solve the first subproblem via reformulation, the primal dual optimization theory, and a careful payment design, and solve the second subproblem via standard solvers. We have rigorously proved that our approach finishes in polynomial time, achieves truthfulness and individual rationality economically, and leads to a parameterized competitive ratio for the long-term social cost. Through extensive evaluations using real-world data traces, we have validated the superior practical performance of our approach compared to existing algorithms.

Industrial cloud robotics employs cloud computing technology to provide various operational services, such as robotic control modules that enable customized screwing and welding. Service composition technology enables the flexible implementation of complex industrial robotic applications based on the collaboration of multiple industrial cloud robotic services. Most studies considered cloud robotic services with a single robotic manipulator with a fixed function. To utilize the advantages of coarse-grained services encapsulated by multi-functional robots, manipulators, and control applications, a multi-granularity service composition method is introduced considering the multi-functional resources and capabilities of the cloud robotic services. Then a quality-of-service-aware multi-granularity robotic service composition model is built to evaluate the composition solution. Furthermore, a multi-granularity robotic service matching strategy is proposed according to the matching constraints of coarse-grained services. Six representative multi-objective evolutionary algorithms are adopted to optimize five quality-of-service attributes of the composite service simultaneously. Experiments demonstrate that the proposed multi-granularity robotic service composition method can remarkably improve the quality of robotic composite services for complex manufacturing tasks by utilizing coarse-grained services in addition to fine-grained services. The performances of six multi-objective evolutionary algorithms are compared to determine the most suitable algorithm for the multi-granularity robotic service composition problem.

Service composition is a core issue of cloud manufacturing (CMfg) to integrate distributed manufacturing services for customised manufacturing tasks. Existing studies focus on the quality of service (QoS) in composition by assuming that each service is independent with each other. However, the correlation between services determines whether a composition is feasible in practice and is a primary factor of its QoS. This paper considers two typical correlations, composability-oriented correlation and quality-oriented correlation. The composability-oriented correlation is modelled as a group of constraints to decide whether a solution is feasible. The influence of the quality-oriented correlation between two services on the overall QoS of a composition is quantified by a discount percentage based on their correlation degrees. A mathematical model of correlation-aware service composition is then established. To solve this problem, a many-objective memetic algorithm termed HypE-C (Hypervolume Estimation Algorithm for Multiobjective Optimisation involving Correlation) is designed. Three correlation-based local search strategies are established in the frame of HypE (Hypervolume Estimation Algorithm for Multiobjective Optimisation) to achieve better trade-off among multiple conflicting QoS criteria. Experiments demonstrate the effectiveness of the proposed algorithm HypE-C compared with five many-objective algorithms on eliminating infeasible search space and providing high QoS service composition solutions.

Virtualization is a crucial technology of cloud computing to enable the flexible use of a significant amount of distributed computing services on a pay-as-you-go basis. As the service demand continuingly increases to a global scale, efficient virtual machine consolidation becomes more and more imperative. Existing heuristic algorithms targeted mostly at minimizing either the rate of service level agreement violations or the energy consumption of the cloud. However, the communication overhead among different virtual machines and the decision time of virtual machine consolidation are rarely considered. To reduce both the over-utilized nodes and the under-utilized nodes with the consideration of migration cost, communication overhead, and energy consumption, this paper presents a new iterative budget algorithm in which a budget heuristic and a multi-stage selection strategy are designed to find suitable migration objects and targets simultaneously. Experiments show that the proposed algorithm provides a substantial improvement over other typical heuristics and metaheuristic algorithms in reducing the energy consumption, the number of migrated virtual machines, the overall communication overhead, as well as the decision time.

In the multi-objective optimization discipline (MOD), simultaneous optimization with four or more objectives is referred to as many-objective optimization. Compared with the two-objective and three-objective optimization, many-objective optimization brings a series of new challenges, such as the deterioration of the global search ability of an optimization algorithm, the difficulty of the visualization of Pareto solutions, and the increase of the computational burden. To address these challenges, in this study, an efficient many-objective optimization system was proposed, and this system was utilized to improve the aerodynamics of a Vista-class cruise ship at four crucial wind angles. In the process of design optimization, a parametric model with eight design variables was selected as the initial ship form. The uniform design (UD) sampling technique was employed to design a group of transformed ship forms. The Reynolds-averaged Navier-Stokes (RANS) solver was used to evaluate the aerodynamics of transformed ship forms, and the nearest neighbor mesh (NNM) interpolation method was utilized for the initialization process of each numerical calculation to reduce the computational cost of a single simulation. With the numerical results of all transformed ship forms, four aerodynamic surrogate models were established, using a combined method based on a particle swarm optimization and a radial basis neural network (PSO-RBFNN), to replace large-scale numerical simulation. In addition, the Sobol’ method was introduced to conduct the sensitivity analysis of the design variables. A series of mature genetic algorithms (GAs) were applied for the single-point, two-point, and four-point optimization of the aerodynamics of a cruise ship in sequence. Additionally, the optimal ship form was selected from the Pareto solutions of the four-point optimization based on the quantitative results of the analytical hierarchy process (AHP). Eventually, the dedicated experimental results of the optimized ship form showed that the aerodynamics at the four wind angles were improved together, confirming the effectiveness of the many-objective optimization system.

Due to the high superstructures and complex configurations of luxury cruise ships, the impacts of wind loads on the ships must be analyzed during design. In this study, the aerodynamic performance of a Vista-class cruise ship was investigated through wind tunnel measurements, empirical formulae and numerical calculations. Wind tunnel experiments were conducted in a simulated atmospheric boundary layer wind field, and the measured data exhibited reliable wind load characteristics of the cruise ship. Three empirical formulae were employed to estimate the wind loads on the cruise ship. The comparison between the estimated results and the experimental data indicated that the general characteristics of the wind loads on generic ship types summarized by empirical formulae were not consistent with those on the Vista-class cruise ship. Numerical simulations based on the Reynolds-averaged Navier-Stokes (RANS) equations were performed under the same conditions as those of the wind tunnel experiments. Mesh independence and Reynolds number independence analyses were carried out. The numerical results of the wind loads matched well with the experimental data. Valuable flow field details captured by the numerical calculations were also extracted to analyze the flow characteristics around the cruise ship. The results demonstrated that the Vista-class cruise ship is a typical blunt body. The wind loads on the cruise ship were sensitive to the Reynolds number. The reversed-flow zones around the ship were primarily distributed over the front/back sides of the ship, the top deck, and the ship's leeward side. These reversed-flow zones were mostly induced by the flow separation at the corners of the bow and aft decks and the sharp edges of the top deck. The analysis of the flow characteristics around the ship can provide guidance for future optimization of the aerodynamic performance of the Vista-class cruise ship.

With the development of technologies, such as big data, cloud computing, and the Internet of Things (IoT), digital twin is being applied in industry as a precision simulation technology from concept to practice. Further, simulation plays a very important role in the healthcare field, especially in research on medical pathway planning, medical resource allocation, medical activity prediction, etc. By combining digital twin and healthcare, there will be a new and efficient way to provide more accurate and fast services for elderly healthcare. However, how to achieve personal health management throughout the entire lifecycle of elderly patients, and how to converge the medical physical world and the virtual world to realize real smart healthcare, are still two key challenges in the era of precision medicine. In this paper, a framework of the cloud healthcare system is proposed based on digital twin healthcare (CloudDTH). This is a novel, generalized, and extensible framework in the cloud environment for monitoring, diagnosing and predicting aspects of the health of individuals using, for example, wearable medical devices, toward the goal of personal health management, especially for the elderly. CloudDTH aims to achieve interaction and convergence between medical physical and virtual spaces. Accordingly, a novel concept of digital twin healthcare (DTH) is proposed and discussed, and a DTH model is implemented. Next, a reference framework of CloudDTH based on DTH is constructed, and its key enabling technologies are explored. Finally, the feasibility of some application scenarios and a case study for real-time supervision are demonstrated.

In order to solve the large-scale model service composition problem in the cloud of simulation, this paper proposes a simulation model service composition framework which considers the characteristics of the cloud. This simulation model service composition framework adopts an ontology-based simulation model service description strategy (MSDS). Based on MSDS, the composite service composed of several model services with complex topology connection relationships is generated by the Input/Output semantic connection strength and simulation capability. A contrast experiment is conducted for the empirical verification.

The present paper presents a numerical investigation on the potential of wind-assisted propulsion for merchant ships. In particular, a KVLCC2M hull was equipped with a set of wingsails inspired from those used in the 34th America׳s Cup. The combined thrust due to the propeller and the wingsails required to achieve a given cruising speed was computed by solving the equations of motion. For every wind direction, the wingsail trim was optimised with a genetic algorithm in order to minimise the thrust of the propeller. The aerodynamic forces and moments due to the hull and the wingsails were computed with Reynolds-averaged Navier–Stokes simulations, while the hydrodynamic forces on the hull and rudder were computed by adapting formulations developed for manoeuvrability applications. It was found that the aerodynamic efficiency of the wingsails is critical in order to gain a meaningful thrust contribution. The propeller thrust was decreased by about 10% when sailing crosswind, and the maximum benefit was achieved by sailing at low speed in strong wind conditions. The oil saving was found to be particularly sensitive to the wingsail aspect ratio, suggesting that an efficient wingsail should employ several tall wingsails rather than a few short and larger wingsails.

With the development of cloud computing, more and more virtual computing resources are provided as dynamic services to achieve green and agile computing for different kinds of projects. Virtual machine migration, as one of the main technologies, has brought huge influences on the executive and collaborative efficiency of virtual machines. However, the existing research focuses primarily on management and scheduling of virtual machines. And several studies concerning virtual machine migration in service-oriented computing still have great limitations. Especially, with the expansion of the scale of collaborative computing tasks, the complexity, as well as the number of virtual machines increased significantly. When the pre-allocation of virtual machines becomes unsuitable due to some resource failures or long waiting queue, migration of virtual machine becomes imperative. Considering the influence of initial allocation, this paper proposes a new parallel plant collaborative growth algorithm (namely Parallel-PCGA) which combines efficient plant growth optimisation operator and ring topology-based parallelisation. It brings an effective balance in solution time and accuracy. Experimental results prove that Parallel-PCGA shows high performance in large-scale virtual machine migration in cloud computing.

Applying Portfolio management into product development would be helpful to utilize and manage overall resources effectively and improve production efficiency. In this paper, a three-dimensional visualization method of project information is proposed for the display, query, analysis and decision-making of project portfolio information. The paper presents the application of the method in the process of project portfolio management of product development and introduces the implementation process of a real project portfolio management.

根据大型复杂装备研发周期长、耗资大、风险大的特点,引入前馈控制机制,建立了大型复杂装备研发项目成本控制专家系统的体系结构,同时以该体系结构为主体,给出专家系统的知识表示及推理机的一种实现方法,最终使得知识库和推理机得以独立实现.实现后的系统以知识库和专家系统相结合的方式对装备研发成本进行实时监控和预测,对于输入的问题,系统经过推理后给出相应的参考建议,为成本管理智能化提供可行依据.由于知识库和推理机是相对独立实现,因此该系统相对传统专家系统来说具有更好的可维护性和可靠性.