IET Collaborative Intelligent Manufacturing最新文献

In the realm of customised production modes, where dynamic disturbances are frequent, synchronised operations between production and logistics in cellular assembly systems play a pivotal role in swiftly responding to rapidly evolving personalised demands. The primary challenge lies in achieving efficient synchronisation of production and logistics amidst intricate operational relationships. This study proposes a production-logistics synchronisation mechanism and method of cellular assembly systems based on digital-twins and Out-of-Order execution. This mechanism enables real-time monitoring of operational processes and robust production and logistics operations, by dynamically adjusting the order of instructions completion based on the executability status and priority of the production and logistics instructions. Consequently, the sequence of job instructions is optimised. Finally, the effectiveness of this approach is substantiated through experiments, establishing it as a viable solution for synchronised production and logistics operations in cellular assembly systems.

In recent years, significant advancements in digital information management and new capabilities within Industry 4.0/5.0 systems have transformed production systems, enabling mass customisation as a new realistic paradigm. Additive manufacturing (AM), or 3D printing, represents a revolutionary approach by allowing the creation of highly personalised products without significantly increasing costs or production time. Efficient utilisation of AM resources requires effective production planning and management, particularly in scheduling production orders, which involves complex nesting logic due to the nonidentical nature of the pieces produced. This work aims to generate actionable knowledge for practitioners, enhancing their ability to understand and effectively tackle these challenges. To achieve this, various deterministic heuristics are proposed to solve the nesting/batching process, and their impact on the quality of final scheduling and computational time is analysed. Real datasets are used to evaluate these strategies, solving larger-sized problems than those previously addressed, to assess resolution capacity. This approach allows for practical rules (easily assimilable by practitioners) to be derived, which ultimately enhance the efficiency of AM systems. The results demonstrate that generating heterogeneous builds—distinct in average heights or volumes—not only improves makespan values by approximately 2%, but also, significantly accelerates the scheduling optimisation process. For the largest instances, computational time is reduced from over 1100 s to just 22 s, representing a remarkable 184% reduction. The underlying intuition for this drastic CPU time reduction is that heterogeneous builds benefit MILP solvers by tightening relaxed solutions; that is, fractional values for binary variables tend to align more closely with the final optimal values.

Lucchese, A., Padovano, A. and Facchini, F. (2025), Comprehensive Systematic Literature Review on Cognitive Workload: Trends on Methods, Technologies and Case Studies. IET Collab. Intell. Manuf., 7: e70025. https://doi.org/10.1049/cim2.70025.

In the previously published version of this article, an earlier version of the manuscript was mistakenly published.

We have now corrected the article to reflect the final updated version as intended by the authors. The article has now been corrected online.

We apologise for this error.

The industrial Internet of Things (IIoT) is revolutionising manufacturing and production of small and medium enterprises (SMEs) by enhancing efficiency and product quality. While developed countries like the USA, UK, Canada, Finland, and Japan have widely adopted IIoT, developing nations such as Bangladesh, India, Pakistan, and Malaysia are still lagging. This study explores IIoT adoption in manufacturing SMEs, emphasising its potential for economic growth despite challenges like budget constraints and skill gaps in developing countries. It presents a novel model based on 17 factors from the TOEI (Technology, Organization, Environment, and Individual) framework to support decision-makers in integrating IIoT technologies. The model’s reliability and validity are confirmed through rigorous testing and a survey of three SMEs. This proposed model serves as a roadmap for SMEs, breaking down complex processes into manageable steps, and providing SMEs with a structured approach.

In human-centric smart manufacturing (HCSM), the robot's dynamic obstacle avoidance function is crucial to ensuring human safety. Unlike the static obstacle avoidance of manipulators or mobile robots, the dynamic obstacle avoidance in mobile manipulators presents challenges such as high-dimensional planning and motion deadlock. In this paper, an adaptive whole-body control approach for dynamic obstacle avoidance of the mobile manipulators for HCSM is proposed. Firstly, an adaptive global path planning method is proposed to reduce planning dimension. Secondly, lateral coupling effect term and nonlinear velocity damping constraints are formulated to alleviate motion deadlock. Then, a whole-body dynamic obstacle avoidance motion controller is presented. Through simulations and real-world experiments, the planning time is reduced by 18.65% on average, and the path length by 15.94%, compared to the global RRT benchmark algorithm. The dynamic obstacle avoidance experiment simulates the obstacle combinations such as pedestrians moving in opposite direction, traversing and forming a circle during the robot operation. The proposed motion controller can adjust robot movement in real time according to the change of its relative distance from obstacles, meanwhile maintaining an average safe distance of 0.45 m from dynamic obstacles. It is assumed that the proposed approach can benefit dynamic human–robot symbiotic manufacturing tasks from more natural and efficient manipulations.

Artificial intelligence is transforming the pharmaceutical sector through improvement in critical processes such as quality assurance (QA). However, in Kenya, technical problems in QA processes, including in-process quality control, equipment maintenance, and visual inspections exist. This paper aims to shed light on the potential of AI in improving pharmaceutical QA in Kenya and challenges associated with its integration. A literature search was thoroughly conducted by retrieving articles from Google Scholar. Articles and policy documents with information relevant to AI applications in QA, optimising pharmaceutical processes, and regulatory compliance in Kenya were reviewed and analysed. AI can improve efficiency and precision in various QA processes including warehousing, equipment maintenance, in-process quality control, and visual inspections, among others. Significant challenges to AI incorporation in QA of Kenya's pharma companies include a lack of technical expertise and understanding of AI outcomes, high implementation costs and fear of losing jobs. There should be strengthened collaborations among government, pharmaceutical manufacturers, AI companies, and researchers to address skill-based barriers and financial challenges.

The randomness and persistence of dynamic disturbances pose significant challenges to resource integration, task allocation, and goal setting within production logistics system. To maintain the optimal operational state of production logistics system over the long term, predictive planning and intervention must occur before disturbances arise, whereas adaptive adjustments are necessary to correct system states after disturbances occur. However, the effective implementation of these control strategies is hindered by several obstacles, such as a lack of comprehensive data and valuable knowledge, which impedes the support for opti-state control (OsC). Fortunately, with the advancements in information technologies such as the IoT and digital twins, it is now possible to collect and process vast amounts of real-time, full-lifecycle big data, thereby enabling more informed optimisation decisions. This paper proposes a digital twin and big data-based opti-state control system (DTBD-OsCS). The architecture integrates big data analytics and service-driven patterns, effectively addressing the aforementioned challenges. Within this framework, both predictive opti-state control (POsC) and adaptive opti-state control (AOsC) strategies are incorporated, along with the development of key technologies for implementing big data analysis. The proposed architecture's effectiveness is demonstrated through application scenarios, and experimental results and findings are thoroughly discussed. The results show that the proposed architecture significantly enhances the efficiency of production logistics systems and effectively reduces the cost impact of disturbances on the system.

This work presents a novel solution for multivariate time series anomaly detection in industrial control systems (ICSs), specifically tailored for resource-constrained environments. At its core, the quantized gated recurrent unit variational autoencoder (Q-GRU-VAE) architecture, a significant evolution from conventional methods, offers an extremely lightweight yet highly effective solution. By integrating gated recurrent units (GRUs) in place of long short-term memory (LSTM) cells within a variational autoencoder (VAE) framework, and employing channel-wise dynamic post-training quantization (DPTQ), this model dramatically reduces hardware resource demands. The proposed solution exhibits performance on par with existing methods on the widely used secure water treatment (SWaT) and water distribution (WADI) benchmarks, while being tailored towards applications where computational resources are limited. This dual achievement of minimal resource consumption and preserved model efficacy paves the way for deploying advanced anomaly detection in resource-constrained environments, marking a significant leap forward in enhancing the resilience and efficiency of ICSs.

Shop scheduling and machine layout are two important aspects of discrete manufacturing. There are strong coupling relationships between them, but they were conducted separately in the past, which significantly limits the production performance improvement of discrete manufacturing. At the same time, in the actual process of workshop production, uncertain events not only often occur but also may make the existing scheduling schemes no longer suitable. To address such issues, the integrated optimisation of shop scheduling and machine layout for discrete manufacturing considering uncertain events is proposed in this paper, where the minimum material handling cost, the maximum space utilisation rate and the minimum production completion time are selected as the optimisation objectives. An improved immune genetic algorithm is designed to solve the corresponding mathematical model efficiently by dual-layer encoding, which is good at global optimisation. Moreover, multistrategy redundancy-aware workshop rescheduling is performed to respond to uncertain events that are regarded as production disturbances. The rationality and superiority of the proposed method are verified by a numerical case study of a discrete manufacturing workshop for wood–plastic composite materials with its integrated optimisation of shop scheduling and machine layout, as well as its rescheduling schemes under machine failures.

This paper focuses on the dynamic flexible job shop scheduling problem with constrained maintenance resources (DFJSP-CMR), a pressing challenge in modern manufacturing systems. As traditional rigid scheduling models fall short in meeting the demands of today's dynamic production environments, there is a growing need for intelligent approaches that can seamlessly integrate production scheduling and maintenance planning under resource limitations. To tackle this challenge, we propose a novel hybrid algorithm aimed at minimising makespan while addressing machine deterioration, unexpected failures and constrained maintenance resources. The core of our approach is a deep Q-network with maintenance insertion algorithm (DQN-MI) specifically designed for efficient maintenance scheduling. The algorithm features a 5×3 action space, constructed as compound rules, along with a reward structure that balances machine utilisation efficiency with effective maintenance operations. Extensive computational experiments conducted on diverse problem instances demonstrate that DQN-MI delivers superior performance, further validating the effectiveness and versatility of the proposed method in addressing complex scheduling challenges while maintaining the stability and reliability of manufacturing systems. This research contributes to the advancement of intelligent manufacturing by presenting a robust and practical solution for the integrated management of production scheduling and maintenance planning.