High-temperature robots face dual challenges of thermal protection and load-bearing under extreme conditions. Traditional thermo-mechanical coupled structures, due to their single-function design, fail to meet the demands of lightweighting and multi-functional integration. Bionic structures and triply periodic minimal surface (TPMS) structures each exhibit excellent mechanical and thermal performance, rendering them promising solutions to this challenge. However, integrate simulation-informed and data-driven methods to enable collaborative design of these two types of structures and provide precise guidance for configuration optimization remains a critical scientific challenge for their reliable application. This study proposes a simulation-informed and data-driven collaborative optimization method that combines deep learning and physical simulation to construct a predictive model, and efficiently establishing a mapping relationship between structural parameters and performance responses. Simulation and experimental results show that the optimised structure achieves a 14.25% improvement in thermal shielding capacity, and a 44.85% increase in load-bearing capacity, significantly verifying the effectiveness of the proposed method. The proposed bionic–TPMS composite structure exhibits excellent thermo-mechanical coupled performance and holds promise for application in thermo-mechanical system design, offering new insights and theoretical support for the engineering application of high-temperature robots in extreme environments.
Surface roughness in Material Extrusion Additive Manufacturing varies across a part and is difficult to anticipate during process planning because it depends on both printing parameters and local surface inclination, which governs the staircase effect. A data-driven framework is presented to predict the arithmetic mean roughness (Ra) prior to fabrication using process parameters and surface angle. A structured experimental dataset was created using a three-level Box-Behnken design: 87 specimens were printed, each with multiple planar faces spanning different inclination angles, yielding 1566 Ra measurements acquired with a contact profilometer. A multilayer perceptron regressor was trained to capture nonlinear relationships between manufacturing conditions, inclination, and Ra. To mitigate limited experimental data, a conditional generative adversarial network was used to generate additional condition-specific tabular samples, thereby improving predictive performance. Model performance was assessed on a hold-out test set. A web-based decision-support interface was also developed to enable interactive process planning by loading a 3D model, specifying printing parameters, and adjusting the part's orientation. The system computes face-wise inclination from the model geometry and visualizes predicted Ra as an interactive colormap over the surface, enabling rapid identification of regions prone to high roughness and immediate comparison of parameter and orientation choices.
Internal porosity remains a critical defect mode in additively manufactured components, compromising structural performance and limiting industrial adoption. Automated defect detection methods exist but lack interpretability, preventing engineers from understanding the physical basis of criticality predictions. This study presents an explainable computer vision framework for pore detection and criticality assessment in three-dimensional tomographic volumes. Sequential grayscale slices were reconstructed into volumetric datasets, and intensity-based thresholding with connected component analysis identified 500 individual pores. Each pore was characterized using geometric descriptors including size, aspect ratio, extent, and spatial position relative to the specimen boundary. A pore interaction network was constructed using percentile-based Euclidean distance criteria, yielding 24,950 inter-pore connections. Machine learning models predicted pore criticality scores from extracted features, and SHAP analysis quantified individual feature contributions. Results demonstrate that normalized surface distance dominates model predictions, contributing more than an order of magnitude greater importance than all other descriptors. Pore size provides minimal influence, while geometric parameters show negligible impact. The strong inverse relationship between surface proximity and criticality reveals boundary-driven failure mechanisms. This interpretable framework enables transparent defect assessment and provides actionable insights for process optimization and quality control in additive manufacturing.
Manuel Barusco, Francesco Borsatti, Youssef Ben Khalifa
et al.
Semiconductor manufacturing is a complex, multistage process. Automated visual inspection of Scanning Electron Microscope (SEM) images is indispensable for minimizing equipment downtime and containing costs. Most previous research considers supervised approaches, assuming a sufficient number of anomalously labeled samples. On the contrary, Visual Anomaly Detection (VAD), an emerging research domain, focuses on unsupervised learning, avoiding the costly defect collection phase while providing explanations of the predictions. We introduce a benchmark for VAD in the semiconductor domain by leveraging the MIIC dataset. Our results demonstrate the efficacy of modern VAD approaches in this field.
Yi-Ping Chen, Vispi Karkaria, Ying-Kuan Tsai
et al.
Digital Twin -- a virtual replica of a physical system enabling real-time monitoring, model updating, prediction, and decision-making -- combined with recent advances in machine learning, offers new opportunities for proactive control strategies in autonomous manufacturing. However, achieving real-time decision-making with Digital Twins requires efficient optimization driven by accurate predictions of highly nonlinear manufacturing systems. This paper presents a simultaneous multi-step Model Predictive Control (MPC) framework for real-time decision-making, using a multivariate deep neural network, named Time-Series Dense Encoder (TiDE), as the surrogate model. Unlike conventional MPC models which only provide one-step ahead prediction, TiDE is capable of predicting future states within the prediction horizon in one shot (multi-step), significantly accelerating the MPC. Using Directed Energy Deposition (DED) additive manufacturing as a case study, we demonstrate the effectiveness of the proposed MPC in achieving melt pool temperature tracking to ensure part quality, while reducing porosity defects by regulating laser power to maintain melt pool depth constraints. In this work, we first show that TiDE is capable of accurately predicting melt pool temperature and depth. Second, we demonstrate that the proposed MPC achieves precise temperature tracking while satisfying melt pool depth constraints within a targeted dilution range (10\%-30\%), reducing potential porosity defects. Compared to PID controller, the MPC results in smoother and less fluctuating laser power profiles with competitive or superior melt pool temperature control performance. This demonstrates the MPC's proactive control capabilities, leveraging time-series prediction and real-time optimization, positioning it as a powerful tool for future Digital Twin applications and real-time process optimization in manufacturing.
In this work we investigate the ability of large language models to predict additive manufacturing defect regimes given a set of process parameter inputs. For this task we utilize a process parameter defect dataset to fine-tune a collection of models, titled AdditiveLLM, for the purpose of predicting potential defect regimes including Keyholing, Lack of Fusion, and Balling. We compare different methods of input formatting in order to gauge the model's performance to correctly predict defect regimes on our sparse Baseline dataset and our natural language Prompt dataset. The model displays robust predictive capability, achieving an accuracy of 93\% when asked to provide the defect regimes associated with a set of process parameters. The incorporation of natural language input further simplifies the task of process parameters selection, enabling users to identify optimal settings specific to their build.
Ahmet Bilal Arıkan, Şener Özönder, Mustafa Taha Koçyiğit
et al.
We present an integrated machine learning framework that transforms how manufacturing cost is estimated from 2D engineering drawings. Unlike traditional quotation workflows that require labor-intensive process planning, our approach about 200 geometric and statistical descriptors directly from 13,684 DWG drawings of automotive suspension and steering parts spanning 24 product groups. Gradient-boosted decision tree models (XGBoost, CatBoost, LightGBM) trained on these features achieve nearly 10% mean absolute percentage error across groups, demonstrating robust scalability beyond part-specific heuristics. By coupling cost prediction with explainability tools such as SHAP, the framework identifies geometric design drivers including rotated dimension maxima, arc statistics and divergence metrics, offering actionable insights for cost-aware design. This end-to-end CAD-to-cost pipeline shortens quotation lead times, ensures consistent and transparent cost assessments across part families and provides a deployable pathway toward real-time, ERP-integrated decision support in Industry 4.0 manufacturing environments.
The re-entrant flow with an unpredictable nature of arrival would apparently harm production plans and schedules in flow type of shops. The re-entrant flow with varied arrival frequencies in rotor blade manufacturing is quite complicated and results in disproportionate workloads. Hence, an attempt has been made to study the significant influence of disproportionate workloads and research on an innovative order release method to enhance performance. The manufacturing process was observed thoroughly to incorporate the uncertain events that cause disturbance in the production. A simulation model was developed on a discrete event simulation platform by analysing problem phenomena right from the conceptualization phase. The model has been verified and validated to ensure the accuracy. The model was subjected to 288 experiments representing different scenarios that a flow shop undergoes in reality. The factors considered in the experimentation were re-entrant frequency, re-entrant proportions, order release methods and priority dispatching rules. A refined load release policy for disproportionate loads has been proposed to judge its effectiveness in terms of profit computation by comparing it with other relevant policies. Results of the experiment revealed that the order release methods contribute 95.93% to throughput performance, in addition, the use of the new re-entrant method policy in the above scenario was productive in improving the overall shop performance.
Three-dimensional (3D) printing has been profoundly changing the production mode of traditional industries. However, this technique is usually limited to metre-scale fabrication, which prevents large-scale 3D printing (LS3DP) applications such as the manufacturing of buildings, aircraft, ships, and rockets. LS3DP faces great challenges, particularly, it not only requires confronting problems not yet solved by conventional 3D printing, such as the inability to print functional structures due to limitations by single-material manufacturing, but also needs to overcome the size effect limitation of large-scale printing. Here, we systematically review the state of the art in the integration of materials and technologies in LS3DP. We also demonstrate some disruptive engineering cases of LS3DP in the field of construction. The challenges and strategies for overcoming size constraints to achieve LS3DP of functional structures are discussed, including multifunctional 3D printing processes from nano- to large-scale and large-scale 4D printing processes, diverse printable materials and sustainable structures, horizontal and vertical size-independent printers, collaborative and intelligent control of the entire process, and extreme environment printing. These strategies can provide tremendous opportunities for the fully automated, intelligent, and unmanned production of these different material megastructures and internal multiscale multifunctional components such as buildings/structures, aerospace vehicles, and marine equipment.
One of the most promising use-cases for machine learning in industrial manufacturing is the early detection of defective products using a quality control system. Such a system can save costs and reduces human errors due to the monotonous nature of visual inspections. Today, a rich body of research exists which employs machine learning methods to identify rare defective products in unbalanced visual quality control datasets. These methods typically rely on two components: A visual backbone to capture the features of the input image and an anomaly detection algorithm that decides if these features are within an expected distribution. With the rise of transformer architecture as visual backbones of choice, there exists now a great variety of different combinations of these two components, ranging all along the trade-off between detection quality and inference time. Facing this variety, practitioners in the field often have to spend a considerable amount of time on researching the right combination for their use-case at hand. Our contribution is to help practitioners with this choice by reviewing and evaluating current vision transformer models together with anomaly detection methods. For this, we chose SotA models of both disciplines, combined them and evaluated them towards the goal of having small, fast and efficient anomaly detection models suitable for industrial manufacturing. We evaluated the results of our experiments on the well-known MVTecAD and BTAD datasets. Moreover, we give guidelines for choosing a suitable model architecture for a quality control system in practice, considering given use-case and hardware constraints.
Kaidong Song, A. N. M. Tanvir, Md Omarsany Bappy
et al.
Thermoelectric materials, which can convert waste heat into electricity or act as solid-state Peltier coolers, are emerging as key technologies to address global energy shortages and environmental sustainability. However, discovering materials with high thermoelectric conversion efficiency is a complex and slow process. The emerging field of high-throughput material discovery demonstrates its potential to accelerate the development of new thermoelectric materials combining high efficiency and low cost. The synergistic integration of high-throughput material processing and characterization techniques with machine learning algorithms can form an efficient closed-loop process to generate and analyze broad data sets to discover new thermoelectric materials with unprecedented performances. Meanwhile, the recent development of advanced manufacturing methods provides exciting opportunities to realize scalable, low-cost, and energy-efficient fabrication of thermoelectric devices. This review provides an overview of recent advances in discovering thermoelectric materials using high-throughput methods, including processing, characterization, and screening. Advanced manufacturing methods of thermoelectric devices are also introduced to realize the broad impacts of thermoelectric materials in power generation and solid-state cooling. In the end, this paper also discusses the future research prospects and directions.
S. Mohadeseh Taheri-Mousavi, Michael Xu, Florian Hengsbach
et al.
Additively manufactured (AM) aluminum alloys with high strength and thermal stability have broad applications in turbine engines, vacuum pumps, heat exchangers, and many other industrial systems. Employing precipitates with an L1$_2$ structure to block dislocation motions is a widespread strategy to strengthen aluminum. However, to achieve high strength, a high volume fraction of small precipitates is required, and these characteristics are generally mutually exclusive. Here, we show that for certain compositions of Al alloys, L1$_2$ phases initially precipitate as sub-micron metastable ternary phases under the rapid solidification conditions of powder bed AM, yet the subsequent L1$_2$ phases that precipitate during heat treatment of the sample remain at the nanoscale, imparting high strength. For strength to be retained at elevated temperature, these nanoprecipitates must have low coarsening rates. To inversely design the composition of an alloy to have these target microstructural features, we used hybrid calculation of phase diagram (CALPHAD)-based integrated computational materials engineering (ICME) and Bayesian optimization techniques. We tested our approach by designing an Al-Er-Zr-Y-Yb-Ni model alloy, and the selected composition was manufactured in powder form as AM feedstock. The strength of specimens manufactured via laser powder bed fusion (LPBF) from the designed composition is comparable to that of wrought Al 7075, yet without cracking that occurs upon LPBF of Al 7075. After high-temperature (400$^\circ$C) aging the designed alloy is 50% stronger than the strongest known benchmark printable Al alloy.
A digital twin (DT), with the components of a physics-based model, a data-driven model, and a machine learning (ML) enabled efficient surrogate, behaves as a virtual twin of the real-world physical process. In terms of Laser Powder Bed Fusion (L-PBF) based additive manufacturing (AM), a DT can predict the current and future states of the melt pool and the resulting defects corresponding to the input laser parameters, evolve itself by assimilating in-situ sensor data, and optimize the laser parameters to mitigate defect formation. In this paper, we present a deep neural operator enabled computational framework of the DT for closed-loop feedback control of the L-PBF process. This is accomplished by building a high-fidelity computational model to accurately represent the melt pool states, an efficient surrogate model to approximate the melt pool solution field, followed by an physics-based procedure to extract information from the computed melt pool simulation that can further be correlated to the defect quantities of interest (e.g., surface roughness). In particular, we leverage the data generated from the high-fidelity physics-based model and train a series of Fourier neural operator (FNO) based ML models to effectively learn the relation between the input laser parameters and the corresponding full temperature field of the melt pool. Subsequently, a set of physics-informed variables such as the melt pool dimensions and the peak temperature can be extracted to compute the resulting defects. An optimization algorithm is then exercised to control laser input and minimize defects. On the other hand, the constructed DT can also evolve with the physical twin via offline finetuning and online material calibration. Finally, a probabilistic framework is adopted for uncertainty quantification. The developed DT is envisioned to guide the AM process and facilitate high-quality manufacturing.
Sudeshna Roy, Hongyi Xiao, Vasileios Angelidakis
et al.
The thermal and mechanical behaviors of powders are important for various additive manufacturing technologies. For powder bed fusion, capturing the temperature profile and the packing structure of the powders prior to melting is challenging due to both the various pathways of heat transfer and the complicated properties of powder system. Furthermore, these two effects can be coupled due to the temperature dependence of particle properties. This study addresses this challenge using a discrete element model that simulates non-spherical particles with thermal properties in powder spreading. Thermal conduction and radiation are introduced to a multi-sphere particle formulation for capturing the heat transfer among irregular-shaped powders, which have temperature-dependent elastic properties. The model is utilized to simulate the spreading of pre-heated PA12 powder through a hot substrate representing the part under manufacturing. Differences in the temperature profiles were found in the spreading cases with different particle shapes, spreading speed, and temperature dependence of the elastic moduli. The temperature of particles below the spreading blade is found to be dependent on the kinematics of the heap of particles in front, which eventually is influenced by the temperature-dependent properties of the particles.
Widya Prananta, Vitradesie Noekent, Angga Pandu Wijaya
et al.
Due to growing competition, customer retention has become a big problem in many service companies. Within a conceptual model for consumer switching intention, this article examines the influence of customer experience, customer satisfaction, and switching intention to green products, all of which are controlled by online information. Even though scholars have researched consumer switching intentions and the elements that influence them, the complex structural processes that minimize the chance of switching intentions in higher education institutions have remained unstudied. This paper addresses the role of customer experience and customer satisfaction to switching intentions moderated by online information. Data is collected through a questionnaire survey. This study employs purposive sampling to obtain respondents (n = 135), with the criteria, students at Universitas Negeri Semarang, a Conservatory University, who wish to switch to green products. Empirical findings support the proposed model and hypotheses, demonstrating that (1) customer satisfaction is negatively related to switching intention, and (2) The online information factor further strengthens the relationship between customer satisfaction and switching intention. The findings of this study provide a unified understanding of the structural relationships that contribute to increased green switching intention to the development of disconfirmation theories in the higher-education context. Implementing a green campus within Universitas Negeri Semarang, campus communities are encouraged to switch to environmentally friendly products to support green campus policies.
Production management. Operations management, Management. Industrial management
David J. Castro Rodriguez, Eleonora Pilone, Gianfranco Camuncoli
et al.
NaTech accidents are a class of cascading events that occur when natural and technological hazards collide. In the process industry, where multi-hazard substances are used in large quantities, failures due to natural events can bring simultaneously or sequentially events of acute toxicity, fire, and explosion, which might impact the population and the environment, also provoking economical losses.
The risk analysis methodology used by the Seveso industry often resulted in scenarios related to NaTech events being excluded due to their low probability. However, the increasing impacts of climate change may lead to variations in the recurrence of severe unexpected natural events that will greatly alter the projected frequency of NaTech events. For this reason, it is critical that decision-makers be adequately informed about potential NaTech risks and consider them not only in industrial safety reports but also in the provisions of emergency and city plans. In this paper, a planning tool is used to assess NaTech risk at a Seveso facility that manufactures lubricating oil additives.
A validated method was used to cross the information among the vulnerable industrial items, the typical damage modes triggered by the natural hazards, and the hazardous substances involved in the plant. The information was extracted from the public inventory of establishments at risk of major accidents connected with dangerous substances, and the safety report that the plant draws up.
The results provide an early warning system to the decision-makers about the NaTech vulnerabilities that threaten both, human health, and the environment, contributing to increasing their awareness and preparedness. Further research is required to integrate this kind of analysis with diverse current methodologies for characterizing NaTech events within territorial and multi-risk approaches.
Chemical engineering, Computer engineering. Computer hardware
Giuseppe Bruni, Sepehr Maleki, Senthil K. Krishnababu
Applications of deep learning to physical simulations such as Computational Fluid Dynamics have recently experienced a surge in interest, and their viability has been demonstrated in different domains. However, due to the highly complex, turbulent, and three-dimensional flows, they have not yet been proven usable for turbomachinery applications. Multistage axial compressors for gas turbine applications represent a remarkably challenging case, due to the high-dimensionality of the regression of the flow field from geometrical and operational variables. This paper demonstrates the development and application of a deep learning framework for predictions of the flow field and aerodynamic performance of multistage axial compressors. A physics-based dimensionality reduction approach unlocks the potential for flow-field predictions, as it re-formulates the regression problem from an unstructured to a structured one, as well as reducing the number of degrees of freedom. Compared to traditional "black-box" surrogate models, it provides explainability to the predictions of the overall performance by identifying the corresponding aerodynamic drivers. The model is applied to manufacturing and build variations, as the associated performance scatter is known to have a significant impact on $CO_2$ emissions, which poses a challenge of great industrial and environmental relevance. The proposed architecture is proven to achieve an accuracy comparable to that of the CFD benchmark, in real-time, for an industrially relevant application. The deployed model is readily integrated within the manufacturing and build process of gas turbines, thus providing the opportunity to analytically assess the impact on performance with actionable and explainable data.
Naufal Dwinanda Narra Putra, Robiyanto Robiyanto, Hans Hananto Andreas
This study was conducted to analyze the performance of the portfolio formed with different asset classes. The instrument used is the consumption sector index with 5 cryptocurrencies. Does the formed portfolio have a better performance than the portfolio that is only formed from the consumption sector index. The type of data in this study uses secondary data in the form of a daily frequency time series with a research period from January 2019 to January 2021. The data in this study used quantitative data. Portfolio performance measurement in this study was measured using the ratio of Sharpe, Treynor, Jensen, Sortino, and Omega. Based on the results of the study, it shows that the performance of the consumption sector index portfolio that is hedged with cryptocurrency produces a higher rate of return in the period during the pandemic than in the period before the pandemic. However, there is 1 crypto that produces negative values in each ratio and research period, namely Tether. Overall, the results of this study can be concluded that adding cryptocurrency to the formation of a portfolio will get a better portfolio performance.
Production management. Operations management, Management. Industrial management
Aloysius Haryono, Tanika Dewi Sofianti, Dena Hendriana
Wamena airport experienced accidents in 2002, 2008, 2009, 2013, 2015, and 2016. All accidents were cargo flights and in approach and landing flight phases. As the Swiss Cheese concept, accident happened when errors penetrated safety defenses’ layers in straight line. Structuring NTSC’s investigations, under HFACS framework to understand the human factor failures type and HFIX strategy to close the failures by applying the recommendations, need to be done in air accident investigation. Eleven aviation experts and practitioners were interviewed in this study, to validate the framework. There were layers without any failures in accident 2008, 2013, and 2016. Accident in 2016 has no recommendation due operators’ safety actions were considered relevant to block failures. Accidents in 2002, 2009, 2013, and 2015 have failure in a layer which intervened by two or more recommendations. There were failures remain open in accident 2002, 2009, 2013, and 2016. Repetitive failure, error or violation of repetitive accidents in 2002, 2009, 2013, 2015, and 2016 is un-stabilized approach and has not been blocked with effective interventions. HFACS and HFIX are useful to framework the accident investigation, preventing similar accident happened in the future.
Production management. Operations management, Business