This study introduces a Digital Twin framework for real-time monitoring and decision-making in automotive strut tower manufacturing via automated high-pressure die casting. As a key enabler of Industry 4.0, the Digital Twin concept integrates physical and digital domains, allowing bi-directional communication for process optimization. The proposed model collects real-time data from sensors in the manufacturing cell and utilizes FlexSim simulation software alongside machine learning algorithms, specifically Artificial Neural Networks (ANN) and Logistic Regression, to classify product outcomes. The ANN achieved high accuracy, supporting rapid, data-driven decision-making. Additionally, FlexSim’s Emulation tool and Open Platform Communications United Architecture (OPC UA) protocol enable seamless communication between the physical and digital systems, ensuring consistent and reliable performance. Results demonstrate the effectiveness of the Digital Twin in enhancing manufacturing efficiency and predictive accuracy, highlighting its potential for broader applications across various manufacturing stages and industries.

In the era of Industry 5.0, manufacturing systems are increasingly required to integrate human-centric principles with real-time adaptability and responsiveness. Mixed-model assembly lines designed to accommodate multiple product variants play a pivotal role in this context, yet must overcome the challenges posed by frequent product changes, diverse human capabilities, and fluctuating operational conditions. Ensuring balanced and efficient operations under these constraints demands a dynamic approach to plan production and system configuration, one that incorporates the variability inherent in human work while maintaining steady throughput. This study proposes a dynamic assembly line rebalancing model that integrates deep reinforcement learning (DRL) with a high-fidelity discrete event simulation (DES). By framing the rebalancing process as a sequential decision problem, the DRL agent continuously adjusts task allocations across manual and automated stations, accommodating stochastic factors such as learning-forgetting effects and shifting production demands. The DES model developed based on a learning factory representing a reconfigurable assembly system ensures realistic modeling of both material flow and human-system interaction, capturing unpaced assembly processes. Comparative experiments demonstrate that the DRL-based strategy outperforms conventional heuristics in productivity, cycle-time reduction, and workload balance without retraining in varied operational scenarios. These findings highlight the framework’s scalability and generalization capacity, as well as its potential to optimize resource utilization and reduce work-in-progress levels. By merging DRL with advanced simulation, this approach enables self-adaptive scheduling that aligns with Industry 5.0 principles. Moreover, it offers a viable pathway toward resilient, human-focused assembly systems capable of real-time, data-driven responsiveness to complex manufacturing dynamics.

For surface quality control, a digital twin model was created and tested. A prediction model with an embedded neuro-fuzzy adaptive inference engine calculates and compares the surface roughness with the required values by utilizing real-time inputs on process parameters, tool wear, acoustic emission, and force signals. Fuzzy logic controls create control commands to modify the machining variables and achieve an appropriate surface quality. Simulation results demonstrate that the developed DT system significantly reduces errors between desired and predicted surface roughness from 11 to 0.8

This research studies high-density polyethylene (HDPE) behavior that is not easily identifiable in a production setting, but is, nevertheless, necessary for consideration when specifying equipment performance for hot plate welding (HPW) applications. Thermal expansion in polymers can depend on several factors, such as the specific type of polymer, its degree of cross-linking, and the temperature range over which the expansion occurs. This physical behavior can affect the process by prolonging the cycle time if the equipment is not sized correctly for the application. This research uses HDPE samples with different surface areas and varying process parameters (force and temperature) to collect data on the sample size change. Experimental results are compared with a recommended industrial guideline of 0.2 to 0.5 MPa of pressure for the Matching stage (stage for surface conformation). Obtained findings indicate that this industrial guideline is very temperature dependent. Data collected in the experiment were used to develop a mathematical model for thermal expansion under different parameters. This study also presents a visual interpretation of HDPE behavior on the effect of temperature if the pressure is increased or decreased.

Industry 4.0 represents a new possibility for manufacturing companies to integrate various digital technologies into their factory floors. These technologies support an interconnected data-driven production, but without an established data foundation at the companies, it can be difficult and expensive to get started on the road towards Industry 4.0. A shift towards new production constellations based on Industry 4.0 will need both new technologies implemented, competencies to handle both data and technologies, and support throughout the organisation of the production companies. This paper suggests a decision-making learning factory aimed towards SMEs; as an artefact to support the learning factory, a mobile IIoT suitcase containing data collected from the SMEs' factories. The data collected were used to help the companies towards building a foundation for datadriven decision-making. An experiment involving 6 SMEs was conducted. The findings show that technology is not a limitation for the SMEs to adopt Industry 4.0, but rather the limited in-house competencies and an understanding of the benefits of implementing industry 4.0 at the organisational level. However, the IIoT pilot project highlights the companies' need for learning-to-learn competencies to adopt new paradigms like industry 4.0. The experiment demonstrated that SMEs see a new aspect of SME production and production planning due to having data-based decision-making.

With the rise in the global population, the challenge of producing more and better-quality productsProduct/ Process Qualification while being profitable becomes increasingly essential. Greenhouse farming (GF) can help the agriculture sector by enabling year-round plant productionProduction regardless of location, climate, and other environ-mental factors. However, this will be realized when they can properly manage their productionProduction processes and limited resources. The lack of accurate and sufficient data is a significant barrier to traditional GF. Digital TwinDigital Twin-Value Stream (DTVS) is an emerging technology that can benefit this industry by providing decision-makers with more precise insight into their operations by employing a Digital TwinDigital Twin at the valueValue stream level. However, to utilize such technology in greenhouses, understanding where they stand is a prerequisite for deciding future actions. The present study aims to provide a method to obtain essential in-formation that will enable DTVS developmentDevelopment by capturing the current capabilities of greenhouses and supporting growers in finding their strategic roadmapRoadmap for employing new technologies. The developed Maturity Assessment ModelMaturity Assessment Model identifies critical core advanced technology dimensions to analyze before proceeding with DTVS developmentDevelopment. A questionnaire to extract needed information, numerical equations to model and quantify the results for better analysis, and an assessment procedure to guide growers to develop and realize a plan for the DTVS implementationImplementation. The results demonstrate that the developed sec-tor-specific maturity model supports transitioning from traditional GF management to real-time monitoring and intelligent decision-making.

Product surface quality which significantly influences a product’s wear resilience and fatigue strength is an important technical characteristic. A frequently used indicator of surface quality, precise mating surface fit, and fatigue life is average surface roughness. Unstable machining leads to an unacceptable surface roughness of the workpieces. Effective surface quality control is significant in improving machining efficiency and reducing costs. A method for controlling the quality of product surfaces based on an offline digital twin and artificial intelligence is developed. To reduce average surface roughness, genetic algorithms are used to optimize machining settings. Actual machining data is used to test and validate the digital twin developed. The simulated results agree with the machining trials results and the root means square error is found to be between 0.128 and 0.135 µm. The developed offline digital twin accurately predicts the surface finish and significantly improves manufacturing and production intelligence.

The challenge of sustainability rests on the ability of organizations to change their practices to meet the needs of current and future generations. To date, most research on organizational change has focused on how to change within a single organization. However, an increasing number of sustainability challenges require changes across multiple organizations. In this paper, we summarize strategic challenges faced in such a setting and outline a conceptual modeling approach for strategic analysis of alliance-driven solutions. We illustrate our ideas with a case study in digital agriculture, a field particularly relevant to sustainability, and end with the identification of issues for further research.

Innovation and transformative changes in products, manufacturing technologies, business strategies, and manufacturing paradigms have profoundly changed the manufacturing systems. In addition to being environmentally, economically socially sustainable, manufacturing systems are increasingly using intelligent technologies to be even more resilient, responsive, and adaptable. A new Adaptive Cognitive Manufacturing Systems (ACMS) paradigm, its drivers, enablers, and characteristics, including cognitive adaptation, is presented. Classification and definitions of four types of adaptability in manufacturing systems are included. Human-centric collaboration of workers and intelligent machines and applications, and the future of work in cognitive adaptive manufacturing systems are outlined. Cognitive Digital Twins (CDT), their features, evolution, and their use to support humans in intelligent, collaborative manufacturing settings are discussed. Industrial applications and case studies are used to illustrate the presented concepts and paradigms. Challenges and future research directions to achieve the ACMS paradigm and implement more intelligent, more adaptive, and sustainable manufacturing systems are presented. The presented novel concepts and technologies make significant contributions to the fast-evolving field of manufacturing systems. This pioneering research sheds light on many important future research topics and provides a road map and motivation for researchers in this field.

Forming product platforms is an effective strategy to offer products variety economically. The short products life span and new variants pose a challenge to designing platforms, which satisfy the changing customer demands. There is a need to design platforms that can adapt to changes. In addition, it is important to account for the partially completed platforms inventory held in each production period which could be utilised in subsequent production periods. An important contribution of this work is the use of combined additive and subtractive manufacturing in customising the product platform by adding and/or removing features to suit the changing product features and demands for different product variants. In this paper, a holistic non-linear model is presented for designing optimal multi-period additive/subtractive product platforms and managing their inventory. The model provides the optimal product platform design for each production period, macro process plans for customisation, the number of each platform stored as inventory, and the product variant platform assignment. The initial model is subsequently linearised to reduce computation time. A gear shaft family of products is used as a demonstration example. The model redesigns the product platform as needed to meet the changing demand of each product variant while minimising the costs.

"Biologicalisation in Manufacturing" is used to integrate biological and bio-inspired processes, principles and resources into manufacturing to handle challenges such as product proliferation. This paper introduces a product variety management methodology utilizing bio-inspired phylogenetics for designing product platforms for the delayed product differentiation using hybrid additive/subtractive manufacturing technologies. The median joining phylogenetic network, an ancestral sequence recognition method used in biology, is applied. This method is used to design product platforms that resemble the common ancestors in nature and determine the process plans for the platform customization that resembles the adaption in nature. A new feature extraction procedure is introduced to extract the additive and subtractive features of a product family. Two case studies, guiding bushings and flanges product families, are used for demonstration. The developed methodology improves the responsiveness and product mix flexibility and decreases the storage costs.

The evolution of manufacturing systems, influenced by changes along four axes - products, technology, business strategies and production paradigms - is presented. Adoption of human-centric decision making in meshed collaboration with intelligent systems is examined. Implications and preparedness for the shift towards more responsive, intelligent adaptive systems are reviewed. Research and industrial use cases are presented. A vision for the new future Adaptive Cognitive Manufacturing System (ACMS) paradigm and its characteristics, drivers and enablers are articulated highlighting the digital and cognitive transformations. Perspectives and insights are offered for future research, education, and work to realize the evolution of manufacturing systems.

Industry 4.0 promotes the utilization of new exponential technologies such as additive manufacturing in responding to different manufacturing challenges. Among these, the integration of additive and subtractive manufacturing technologies can play an important role and be a game changer in manufacturing products. In addition, using product platforms improves the efficiency and responsiveness of manufacturing systems and is considered an enabler of mass customization. In this paper, a model to design multiple platforms that can be customized using additive and subtractive manufacturing to manufacture a product family cost-effectively is proposed. The developed model is used to determine the optimal number of product platforms, each platform design (i.e. its features set), the assignment of each platform to various product variants, and the macro process plans for customizing the platforms while minimizing the overall product family manufacturing cost. The multiple additive/subtractive platforms and their process plans are determined by considering not only the commonality between the product variants but also their various manufacturing cost elements and the customer demand of each variant. The design of multiple product family platforms and their process plans is NP-hard problem. A genetic algorithm-based model is developed to reduce the computational complexity and find optimal or near optimal solution. Two case studies are used to illustrate the developed multiple platform model. The model results were compared with a single platform model in literature and the results demonstrate the multiple platform model superiority in manufacturing product families in lower cost. The use of the developed model enables manufacturing product families cost efficiently and allows manufacturers to manage diversity in products and market demands.

With the advent of industry 4.0 and smart manufacturing, the panorama for factories and businesses is changing at high speed at all operation levels. Hence, it is imperative that human capital keeps up and adapts to meet present and future needs. Professionals should possess breadth and depth in knowledge along with meta-skills to thrive in the rest of the 21st century. This paper discusses how technologies have impacted and shifted the roles of people in the factory and their interactions with smart machines and systems, and the challenges that professionals will face in the future. Initiatives that governments and organizations are rolling out to bridge the skills gap are examined. The desired adaptation of education necessary to produce professionals ready for the fast-changing manufacturing environment is discussed. Recommendations for successful human development, through reskilling and upskilling, in the smart manufacturing context are provided.

ABSTRACT A novel integer programing (IP) model is developed to assign various operations of product variants to candidate machines and select the best place for each machine among the candidate locations. Product variants are considered to have operation sequence flexibility which are optimized for each product variant. The objective is to minimize the total backtracking distance by considering the production volume of each variant. Backtracking occurs when the production flow is upstream. Minimizing backtracking will improve forward production flow and reduces unnecessary distance travelled by parts which can lead to significant reduction in cycle time and improve machine utilization and total system throughput. In the literature, backtracking problem has been addressed in generalized flow line (GFL) problems, and flexible manufacturing systems. There is no work in the literature considering product variants with networked operation sequence as well as machine selection for each operation. A non-linear mathematical model is first formulated, then converted to an equivalent integer programing (IP) model and solved. A family of engine cylinder blocks is used as a case study for demonstration. The obtained results reveal that minimizing backtracking can significantly improve the total throughput and reduce the total traveling distance (33.33% and 27.78%, respectively, in the case study).

Product platforms represent an effective strategy implemented by manufacturers to cope with dynamic market demands, decrease lead-time and delay products differentiation. A decision support system (DSS) for product platforms design and selection in high-variety manufacturing is presented. It applies median-joining phylogenetic networks (MJPN) for the platforms design and phylogenetic tree decomposition for platforms selection by determining the product family phylogenetic network and defines the platforms at various levels of assembly corresponding to different trade-offs between number of platforms (variety) and number of assembly/disassembly tasks (customisation effort). Product platforms are reconfigured and customised to derive final product variants. The phylogenetic tree is decomposed in multiple levels, from the native platforms to the final variants. New Platforms Reconfiguration Index (PRI) and Platforms Customisation Index (PCI) were developed as metrics to evaluate the platforms customisation effort. A case study of a large family of plastic valves is used to demonstrate the DSS application. It shows reduction of 60% in platforms variety and increases in platform customisation assembly/disassembly tasks by only 20% leading to significant production and inventory efficiencies and cost savings. This methodology supports companies in the design and selection of best product platforms for high-variety to reduce cost and delivery time.

Nowadays, manufacturers aim at satisfying diverse and changing customer demand in order to survive in the competitive market. An approach that combines several manufacturing concepts, including product platform formation and hybrid manufacturing, is proposed in order to effectively manage the product variety. A genetic algorithm-based model is introduced to design the optimal or near-optimal platform for large sets of products and features that can be further manufactured by additive and/or subtractive manufacturing to be customized into different product variants. An illustrative example is used to demonstrate the model. The proposed model leads to better management of product proliferation.

The platform strategy has been implemented to efficiently manage the increased variety in products and manufacturing systems domains by achieving their effective and rapid re-configuration. Despite the increased development of platforms research, their back-end issues such as the supply chain and supplier selection have received little attention. In this research, a methodology that integrates the product platform synthesis with the selection of suppliers to form a supplier platform is introduced. The formed supplier platform is a collection of suppliers capable of supplying the components/modules of the product platform. The supplier platform remains unchanged for product generations, and non-platform suppliers are added or removed as needed for producing different product variants in different production periods. The presented co-development methodology consists of three phases. First, co-platforming is used to map the product requirements to the supplier's domain; then an intuitionistic fuzzy TOPSIS method is employed to assign weights to the suppliers according to selected criteria. The suppliers are chosen next and their platform is synthesized. A laptop product family is used to illustrate the developed methodology. The significance of this research is the synthesis of a supplier platform which can be used without change for many product variants and many product generations. Its implementation enables the planning and creation of strategic alliances with the product platform suppliers.

Data is collected from different industrial domains. Organizing that data makes change anticipation more planned and streamlined. This paper introduces a novel holistic model of associating different domains of industrial data. The model establishes a tree graph called cladogram to create a unified classification of data from market segments, product design and manufacturing capabilities and it is expandable beyond these domains. The cladogram is produced by the widely used biological Cladistics analysis, without modification. This approach has a great degree of simplicity without introducing an extra layer of mathematical modelling, while resulting in a data-inclusive graphical representation. A case study of automated and flexible assembly is presented to demonstrate the effectiveness of the model and its simplicity. Model results are significant, since they could reveal associations of the definitions of the objects from different data domains, which were used later in response to future changes in those domains.