Adaptive active decoding and novel disjunctive graph-based improved genetic algorithm for multi-type machine robot cell scheduling in mass customization

ABSTRACT The trends of mass customization and personalized production lead to increased manufacturing complexity, requiring human involvement for flexibility, particularly in High-Mix, Low-Volume (HMLV) assembly. These production systems impose challenges on operators, highlighting the growing importance of providing cognitive support. Digital assembly instructions have emerged as a solution, offering step-by-step guidance to operators. However, creating and maintaining these instructions is labor-intensive, especially in HMLV contexts, due to numerous product variants and frequent design changes. This study introduces a framework for integrating product variability in digital assembly instructions and streamlining the instruction authoring process. A semantic model based on the industrial standard ISA-95 is proposed to integrate engineering information into assembly instructions. Additionally, a methodology is presented that incorporates a 150% workflow for managing instructions for a product family, from which specific variant configurations can be derived. The system suggests relevant instruction content during the authoring process of a new product variant to enhance the reuse of previously written instructions. A methodology to handle engineering changes within the assembly instructions has also been developed to warrant consistency with the product design. Initial testing has demonstrated promising results, including substantial time savings and improved consistency.

As intelligent manufacturing systems increasingly pursue mass customization and rapid market responsiveness, efficient resource allocation has become a critical determinant of production competitiveness. Dynamic resource scheduling in such systems involves complex challenges stemming from highly variable market demand and complex production conditions. This study extends a multi-modal data fusion approach that incorporates deep learning methods to solve dynamic resource scheduling with variable demand scenarios. The approach involves the fusion of sensor, production, and market data using intermediate fusion methods, the adoption of an LSTM with attention model to predict demand, and Double Deep Q-Network (DDQN) methods to control resource scheduling decisions. Simulation outcomes show that the system achieves 82.6% resource utilization and 89.1% on-time delivery performance in nominal scenarios; however, it performs poorly during peak hours, leading to severe delays accounting for 12.8% of total deliveries. The multi-modal fusion approach outperforms single-source methods by 9.6% in prediction accuracy but exhibits limitations in cross-industry generalization (67.9% efficiency retention) and computational efficiency (18-35min for large-scale scenarios). These findings provide realistic insights into the practical deployment of AI-driven scheduling systems in manufacturing environments.

The paradigm shift from mass production to mass customization has empowered users to actively participate in the product design process. Consequently, interactive product customization has emerged as a critical link connecting user preferences with intelligent manufacturing. However, this system faces two major challenges: the increased cognitive load and decision fatigue caused by explicit feedback (e.g., Rating products), and the difficulty in discerning users’ implicit preferences from the vast amount of procedural information generated during interaction. To address these challenges, this study proposes a novel preference analysis method based on multi-graph that incorporates the user’s selective attention process. This method first achieves a unified representation of gaze and selection behaviors through a graph network, thereby establishing a unified representation for analyzing user cognitive patterns. By integrating global and local information from the graph network, we then propose a selective attention centrality metric to effectively quantify the prominence of different design options within the user’s cognitive process. To demonstrate its practical application, we integrated this metric with an interactive genetic algorithm to develop a customization system for Chinese vases. A controlled experiment was then conducted to compare our method against two traditional approaches. Experimental results show that our proposed method offers significant advantages in multiple metrics, including interaction efficiency, algorithm convergence, and user satisfaction. Specifically, the total evaluation time was reduced by 65.45% and 32.09% compared to the traditional interactive genetic algorithm and eye-tracking analysis methods, respectively. This research combines cognitive science theory with graph network analysis methods. It demonstrates the potential of graph networks in decoding complex human behavior information and building efficient human-AI collaboration systems.

Driven by the rapid evolution of flexible electronics, rehabilitation healthcare is shifting toward devices that seamlessly interface with human body. Yet, existing solutions often simply layer flexible sensor units over rigid components, making it difficult to combine high elasticity, mechanical robustness, and trueimperceptibility. Here, we are pioneering a super-tough (∼54.7MPa) and highly stretchable (>400% strain) triboelectric webbing (T-webbing) that overcomes this long-standing trade-off through the synergistic integration of an embedded textured architecture and functional elastic yarns. The T-webbing supports mass customization, exhibits outstanding electrical durability (>100000 cycles), and enables reliable self-powered sensing capability with tunable mechanical properties for diverse rehabilitation tasks. In a proof-of-concept demonstration, the T-webbing is seamlessly integrated into a machine-learning-enabled lower-limb rehabilitation platform, achieving a motion recognition accuracy of 97.9% while enabling seamless one-click data sharing, intuitive human-machine interaction, and real-time remote guidance. By bridging high mechanical resilience with imperceptible wearability, our study offers a brand-new solution for data-driven, high-compliance, home-based rehabilitation within the Internet-of-Things ecosystem-addressing a pressing clinical need for scalable, patient-friendly solutions.

In mass customization, order tasks must strike a balance between batch production and customization, minimizing costs while ensuring fast delivery. Effective collaboration between enterprises and suppliers necessitates careful consideration of order allocation granularity and dynamic resource production capacities to achieve optimal task distribution. This paper introduces a multi-order collaborative allocation model based on a product-process mix, aimed to address the order allocation challenges faced by core manufacturing enterprises in cloud manufacturing under mass customization. First, the limitations of existing order decomposition and resource constraint strategies in the order allocation process are analyzed. A product-process mixed multi-order collaborative allocation model is then established, focusing on minimizing multiple-order costs while accounting for dynamic resource capacity constraints. Next, an improved ant colony algorithm (IACO-a&b) is proposed, tailored to the features of the model. This algorithm enhances the initialization of the ant search solution space by combining the optimal solution searched by the ants, while dynamically adjusts pheromone volatility coefficients and sets path pheromone concentration intervals, so as to avoid preventing the algorithm from precociousness. Finally, the performance of IACO-a&b is compared with several other algorithms, and the proposed order allocation model is evaluated against traditional models in this paper. The experimental results demonstrate that, in large-scale case studies, IACO-a&b improves the best fitness value by 11.89% and increases the fitness excellence rate by 60%. Furthermore, the proposed model reduces the leveling volatility index (LVI) of resources by 50.79% without significantly increasing order costs.

Injection molding is a foundational manufacturing process for the large-scale production of plastic parts and is also applied to rubber, metal, ceramic, glass, wood, and composites. Regular injection molding focuses on minimizing warpage and ensuring dimensional uniformity and high repeatability, as non-uniform material shrinkage in complex mold cavities causes undesirable deformations. In contrast, we reverse this traditional paradigm by leveraging controlled solidification shrinkage to induce purposeful, programmable deformation. We propose 4D injection molding, utilizing localized thermal activation and selective in-mold bonding to create non-uniform temperature and pressure distributions in spatiotemporal dimensions during the molding cycle. This enables the generation of complex, customized geometries from a single mold. Integrated with predictive modeling and multi-objective response optimization, our approach transforms warpage from a defect into a design feature, yielding functional parts with tunable shapes and performance characteristics. This method supports rapid, scalable, and cost-efficient mass customization across various materials. By reimagining shrinkage not as a limitation but as a design opportunity, the method represents a transformative shift in manufacturing science, with promising applications in biomedical, aerospace, and responsive consumer products.

In the context of the increasing need to improve the management efficiency of enterprises that support the implementation of the principles of digital transformation based on the concept of the fifth-generation industry, the relevance of research on the development of appropriate systems in terms of ensuring continuous targeted and sustainable development, customer-centricity and social orientation of production is increasing. Digital twin technology and its multi-agent implementation act as effective means of building enterprise performance management systems. At the same time, the lack of scientific research in this area determines the purpose of the article, which is to develop a product-resource approach to enterprise performance management based on digital twins in the fifth-generation industry. A distinctive feature of the proposed approach developed by the authors is the use of dynamic enterprise performance management technology based on digital twins, which ensures the integration of business processes and resources used at the level of not only one enterprise, but also at the level of network value chains based on a common digital platform of the business ecosystem. The paper analyzes approaches to the intellectualization of enterprise management, on the basis of which the requirements for an enterprise performance management system are formulated, ensuring the solution of interrelated tasks of targeted enterprise development, the formation of flexible value chains, and the rational and sustainable use of enterprise resources. The possibilities and disadvantages of the efficiency management process in EPC class systems are analyzed. The paper substantiates the use of digital twin technology and its multi-agent implementation to build an enterprise performance management system in the context of mass customization and the network nature of value chains in the fifth-generation industry. A process for managing the efficiency of enterprises at all stages of the life cycle based on the technology of digital twins of products and resources has been developed, dynamically ensuring the targeting, adaptability and sustainability of the functioning and development of the enterprise.

Lattice structures manufactured via additive manufacturing (AM) technologies offer substantial potential for application-specific energy absorption solutions across automotive, aerospace, and medical sectors, yet systematic frameworks for translating performance requirements into manufacturable design parameters remain limited. This investigation examines the mechanical behavior and manufacturing repeatability of Face-Centered Cubic (FCC) and Diamond lattice architectures produced through binder jetting (BJ) technology of thermoplastic polyurethane (TPU) across four density levels, with comprehensive characterization of energy absorption performance and failure mechanisms. Uniaxial compression testing was conducted, with detailed analysis of deformation behavior, densification characteristics, and manufacturing consistency. The deformation evolution of the analyzed structures was captured by a digital camera and analyzed through Digital Image Correlation (DIC). Finite element model was also involved to better understand the stress and strain of different structures. Diamond lattice structures demonstrated superior energy absorption performance with 35–45% higher specific energy absorption capacity compared to equivalent-density FCC architectures, exhibiting progressive loading characteristics and extended useful deformation range beyond 60% strain. Manufacturing repeatability analysis revealed coefficient of variation values below 10% for Diamond structures across all mechanical properties, indicating process maturity suitable for mass customization applications. These results validate binder jetting technology for reliable production of customized lattice-based energy absorption components, providing a systematic framework for application-specific lightweight design across multiple industrial sectors.

Additive Manufacturing (Additive Manufacturing (AM), has progressed from a prototyping tool to a disruptive industrial technology, profoundly impacting sectors such as aerospace, automotive, and healthcare. This systematic review rigorously examines the current state-of-the-art in Design for Additive Manufacturing (DfAM), synthesizing its foundational principles, advanced computational strategies, and the overlooked structural discontinuity between design intent and final physical realization. Utilizing a structured search and selection process adhering to the PRISMA framework across major scientific databases, a total of 65 peer-reviewed studies published within the past decade were critically analyzed. The findings elucidate how DfAM fundamentally leverages the intrinsic design freedoms of AM, enabling the realization of intricate geometries, significant component lightweighting, and unprecedented mass customization. Concurrently, this review exposes the technical barriers that continue to impede widespread industrial scalability, notably material anisotropy, reproducibility issues, fragmented digital design workflows and precision metrology systems, such as CMM, 3D scanning and XCT. A key synthesis is provided on advances in topology optimization, lattice structure design, and the burgeoning integration of artificial intelligence and machine learning, with a specific emphasis on their practical implementation and industrial impact. Special consideration is given to sector-specific applications, particularly in biomedical engineering, where DfAM facilitates patient-tailored solutions but still confronts regulatory compliance and biocompatibility hurdles. Critically, this review maps evidence-based trends to expose the fragmented nature of the design-to-part continuum, providing a necessary, unified framework to guide subsequent research toward industrial maturity. This framework is designed to equip engineers, researchers, and practitioners for advancing DfAM practices, securing its evolution as a cornerstone of sustainable and innovative manufacturing within the industry eras.

Abstract Mass customization is transforming global manufacturing paradigms, fueled by increasing consumer demand for individualized products. Traditional manufacturing methods, constrained by inflexible workflows, high setup costs, and long lead times, struggle to efficiently address these personalized demands. In contrast, additive manufacturing (AM), commonly known as 3D printing, presents a flexible, scalable alternative capable of producing complex and customized designs layer by layer with minimal tooling. This paper examines the role of AM in bridging the divide between traditional mass production and personalized manufacturing. By integrating AM into existing production systems, manufacturers can reduce setup times, enable on-demand production, and respond dynamically to market needs. The paper highlights key AM technologies, Material Extrusion, Vat Photopolymerization, Powder Bed Fusion, Binder Jetting, and Directed Energy Deposition, and reviews recent innovations aimed at improving throughput, material performance, resolution, and system integration. Through diverse applications spanning healthcare, consumer goods, and industrial sectors, the potential of AM to deliver mass-customized products is demonstrated, emphasizing its capacity for personalized medical implants, tailored consumer goods, and responsive, distributed manufacturing systems for critical needs. This is further supported by a comparative productivity analysis against injection molding and a review of research efforts focused on enhancing industrial throughput and scalability. Ultimately, this paper presents additive manufacturing as a pivotal technology ins redefining manufacturing paradigms, highlighting both current innovations and future directions essential for fully realizing mass customization at scale.

This study examines planning in mass customization contexts that face challenges, due to high product variety, sparse configuration-level demand, and long supplier lead times. Traditional Configure-to-Order and Assemble-to-Order (CTO/ATO) planning approaches often rely on late procurement and full postponement, leading to high and unstable customer lead times. To address this problem, a lead-time-first planning approach is developed to translate historical demand information into executable planning decisions without relying on finished-goods inventory. The approach operates across three levels: feature-level Component Readiness Tiering for upstream component pre-positioning, segment-level Mix Guardrails to control demand heterogeneity, and configuration-level Top-K partial pre-kitting to exploit demand concentration while preserving flexibility through postponement. The approach stands out because it directly links demand variability metrics to operational readiness thresholds. This link enables structured staging and coverage-based configuration selection. The approach is evaluated using a synthetic dataset representing one year of demand for customized laptops. Performance is assessed using lead-time-oriented indicators, including the 95th percentile customer lead time and instant-start rate. Results show improved responsiveness, with the worst-case customer lead time reduced from 12 days to approximately 7 days and immediate production enabled for a significant share of orders. These improvements are explained by early readiness of high-demand components combined with postponed final differentiation. The approach suits modular CTO and ATO environments with clear demand segments, stable high-volume components, and regular planning cycles.

The manufacturing sector drives industrialisation and contributes substantially to economic growth and employment creation. Despite this, it faces the challenges of diminishing size and lack of competitiveness, mainly due to operational uncertainty. The study developed an approach to managing operational uncertainty using Industry 4.0 and 5.0 technologies. It employed a multimethod quantitative design based on the post-positivist paradigm, with data collected from 22 experts and 262 responses from a manufacturing firms’ survey. The study employed an integrated fuzzy decision-making trial and evaluation laboratory (DEMATEL) with partial least squares structural equation modelling (PLS-SEM) and fuzzy set qualitative comparative analysis (fsQCA). The fuzzy DEMATEL results reveal that growing geopolitical tension, cost-of-living-driven consumer behavioural change, pandemic turbulence, lack of energy stability and security, and the entrenched power of large firms are causal dimensions of operational uncertainty. Industry 4.0 and 5.0 technologies, with capabilities for scenario planning and supply chain integration, flexible production and mass customisation, real-time system and process monitoring and response, root cause analysis, and sustainable solutions, can manage operational uncertainty. These technologies include artificial intelligence (AI), the Internet of Things (IoT), big data analytics, and, to a lesser extent, advanced robotics, blockchain, and augmented and virtual reality (AR/VR). This study advanced configuration theory and a new integrated methodology (fuzzy-DEMATEL-PLS-SEM-fsQCA) to develop solutions for sustained performance during operational uncertainty in manufacturing. This research offers valuable information to advance the subject, make meaningful changes in day-to-day manufacturing operations, and promote practical real-world problem solving.

Modern industries impose numerous challenges on manufacturing technologies and systems due to stringent quality requirements, the adoption of difficult-to-machine materials, high demands on miniaturization, mass customization, high productivity, and sustainability targets [...]

Purpose Mass personalization (MP) presents challenges such as highly individualized orders and fluctuating demand, requiring more flexible supply chains. This study explores supply chain flexibility (SCF) through a collaborative lens and proposes a framework capturing the dynamic, interdependent nature of supply chain relationships, addressing a key empirical gap on how collaboration mechanisms affect SCF in MP contexts. Design/methodology/approach A single case study was conducted of a European bicycle manufacturer with an established MP strategy and SCF experience. Using an abductive approach, we analyzed semi-structured interviews with internal managers, suppliers, and retailers through a three-stage coding process. Findings Based on Cao et al.’s (2010) collaboration framework, nine supply chain collaboration (SCC) mechanisms supporting SCF were identified, including novel MP-specific mechanisms such as rolling forecasts, order adjustment and digital product configurators. These mechanisms are organized into two interrelated flexibility dimensions: spanning flexibility, referring to rapid and accurate information dissemination across supply chain actors, and ecosystem flexibility, representing a firm's capacity to establish adaptive, trust-based and enduring collaborations with key partners. The study highlights the mutually reinforcing nature of these mechanisms in enabling SCF in MP contexts. Originality/value This study is among the first to examine the intersection of MP, SCC and SCF. It contributes a revised model of network-oriented SCF, introduces and categorizes MP-specific SCC mechanisms, and emphasizes the role of ecosystems. Findings offer actionable insights for managers seeking to align SCC with personalization goals.

ABSTRACT Online Product Configurators (PCs) enable collaboration between customers and businesses for personalised codesign; however, mapping customer requirements to precise module-attribute values in apparel, particularly activewear, remains difficult for non-experts. This work positions expert guidance as a virtual assistant (VA) layer for online PCs, estimating the relative importance of candidate attribute values. This paper presents a model that links primary customer requirements to module-attribute values by eliciting expert knowledge and resolving disagreements via a semantic triangular fuzzy transformation. The results demonstrate clearer option selection, reduced trial-and-error in decision making and a unified representation of diverse expert perspectives. Demonstrated using sports bras, the model aligns module-attribute values with customer requirements by reducing fuzzy expert judgements, making it well-suited as the reasoning engine for a VA. This provides an updatable approach to personalised apparel design that does not require domain expertise from users, while offering a reproducible mapping extendable to other modularisable products.

Abstract Orthoses are effective rehabilitation devices that provide support, protection, comfort, and deformity correction, playing a vital role in improving recovery and functional outcomes. Because of substantial individual anatomical variations, higher customization levels are critical for optimal fit and comfort. However, this typically increases cost and production time. To promote mass customization by reducing time and cost while maintaining satisfactory fit, this study proposes an adjustable semi-customized design methodology and develops twelve modular arm-wrist orthoses assembled with adjustable connectors. Analyses and experiments were conducted to evaluate connector adjustability and validate the overall fitting performance. Connector experiments confirmed their ability to achieve the required adjustability range, with deformation between −9.2 mm and +10.5 mm accommodating intra-cluster variations. Computational analyses demonstrated that the semi-customized designs conform effectively to target surfaces through adjustment. Approximately 75% of surface regions exhibited a normal distance below 1.5 mm between target and orthosis after adjustment, indicating successful adaptation—especially considering the conventional 2–5 mm padding allowance. Furthermore, the fitting experiment yielded positive participant feedback, confirming the satisfactory performance of the adjustable semi-customized designs. The proposed methodology thus provides a promising solution to mitigate the excessive time and cost of orthotic manufacturing while maintaining fit quality, contributing to time-efficient and cost-effective mass customization of orthotic devices.

Animal-borne tracking devices (bio-loggers) are established instruments for researching animal behaviour. However, commercial animal trackers are rather standardized and not perfectly adapted to species-specific requirements. Although species-specific solutions are developed, customization effort is high and requires detailed engineering know-how. Furthermore, the development process brings multiple challenges across the process chain and uncertainties for untested species may require iterative refinements in the early design phase. This interdisciplinary study provides a vision of how to enable mass customization of animal trackers through a web-based design platform. The platform involves biologists in engineering processes, makes custom designs accessible to the community, and enhances reusability. Knowledge-based engineering and design automation algorithms are central platform elements, and they automate engineering processes from requirements to the electronic component selection and generation of 3D-printable housing geometries. The animal tracker housings are manufactured using low-cost 3D-printing (additive manufacturing), which offers high flexibility in terms of producible geometry and batch size. Furthermore, this study presents a design automation prototype that implements core functions of the vision to demonstrate the feasibility of automatically generated animal trackers. The software architecture of the design automation prototype and the intermediate algorithm steps are described as open source. To demonstrate the functionality of the design automation prototype, the animal tracker housings of three species are successfully generated and produced. The algorithms take less than 50 seconds to generate the three housings. This demonstrates, how the automation eliminates bottlenecks in the development process and thus greatly reduces efforts for customized animal trackers. The full realisation of the vision can eventually empower biologists to design animal trackers without the involvement of engineers.

Abstract Additive manufacturing (AM) represents a category of manufacturing processes that fabricates parts in a layer-by-layer manner. As such, AM provides unique advantages over conventional manufacturing processes such as the ability to fabricate highly complex geometries, to minimize material waste, and to enable mass customization, while having some limitations, such as high costs and complexities. Advances in artificial intelligence (AI) and machine learning (ML) enable these limitations to be addressed due to the data-rich environment in modern commercial AM machines with multiple sensors. This paper surveys papers that apply AI/ML techniques to the topics of defect detection, AM process surrogate models and their application, generative design, and design for manufacturing in metal AM processes. The approach taken is to introduce these topics, provide a coarse survey, and then discuss specific applications in some depth, rather than to provide a fine-grained, comprehensive survey.

This study investigates how housing can become more affordable and environmentally sustainable by applying circularity principles through mass-customized prefabricated construction. The study proposes that, by using mass-customized concepts and practices and offering buyers interior choices that suit their needs, homes can be made more affordable and sustainable. Sustainability would be achieved through circularity and waste reduction. The discussion of the underlying topic uses a nonreactive desk research methodology. The review of scientific articles and studies covered the following issues: (i) the definition of circularity in housing, (ii) the mass-customized concept in building, and (iii) 3D printing prefabricated home construction. The second stage of evaluating the Affordable Prefab Home project involves examining customized internal modular partitions and prefabricated exterior wall designs. In the discussion, the research identified several barriers to the efficient mass customization of prefabricated housing in the construction industry. Conclusions highlight the importance of adopting mass-customized prefabricated concepts in residential construction.