Industrial demand forecasting is crucial for managing stock levels, financial revenue projections, logistics, and efficient production operations. However, there are notable gaps in the literature, particularly in the research that explores the potential of intelligent monitoring systems using Internet of Things (IoT) devices in combination with industrial production forecasting algorithms. This gap is particularly evident in scenarios characterized by seasonal data and limited historical samples. This research work presents an affordable method for monitoring and forecasting industrial production demand, which involves developing an IoT device and applying deep learning algorithms to an industrial production line. In the proposed methodology, an edge device, known as a datalogger, collects and transmits production data to a cloud server obtained through photoelectric laser beam sensors and the PLC of a filling machine. The data is subsequently retrieved from the cloud database for analysis and production demand forecasting. This work compares the performance of three distinct neural networks–sxLSTM, GRU, and SCINet-in the context of time series forecasting. The evaluation employs metrics such as root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Through this framework, it was possible to achieve to forecast production demand with error rates of 9.08 (RMSE), 7.67 (MAE), and 1.86 (MAPE), due to the SCINet neural network, even in scenarios with seasonal samples and limited historical data from an industrial filling line.
This research presents the MonolayerFFF dataset, a novel addition to Fused Filament Fabrication (FFF) 3D printing image-based datasets, focusing on monolayer parts with unique geometric variations. Unlike existing multi-layered datasets, MonolayerFFF comprises three specific monolayer part conditions: Regular, Defect 1 (with reduced filament deposition), and Defect 2 (featuring filament retraction). This approach offers a different perspective on 3D printing analysis, capturing a range of typical and atypical scenarios. Utilizing the GoogLeNet architecture for Convolutional Neural Network (CNN) classification, the study achieved a notable validation accuracy of 95.81% on the MonolayerFFF dataset, demonstrating its quality and diversity without relying on data augmentation. The high accuracy, coupled with minimal misclassifications as evidenced in both test and validation confusion matrices, underscores the effectiveness of the proposed backbone for feature extraction and classification in identifying subtle differences in part conditions. The study’s outcomes, including the absence of overfitting and aligned loss curves, highlight the model’s capability to generalize from training to unseen data. The MonolayerFFF dataset, with its focus on monolayer parts and detailed variation in conditions, emerges as a significant contribution to advancing CNN analysis in the field of 3D printing. Its potential for real-time monitoring and quality assurance in additive manufacturing promises to enhance the FFF process efficiency and accuracy, making it a valuable tool for researchers in the field.
Computer vision applications demand a significant amount of data for effective training and inference in many computer vision tasks. However, data insufficiency situations usually happen due to multiple reasons, resulting in computational models whose performance is inadequate. Traditional data augmentation techniques are presented to solve this overfitting problem; however, their application is not always possible or desirable. In this context, this paper addresses a different data augmentation technique for classification methods based on adversarial images to reduce the impact of sample imbalance utilizing the Fast Gradient Sign Method (FGSM) with added noise to enhance classifier performance. To validate the method, a set of images was used for the classification of diseases in coffee plants due to the soil’s lack of nutrients. The results showed an improvement in the model performance for this classification in coffee plants proving the validity of the proposed method, which can be used as an alternative to traditional data augmentation methods.
Milling is one, if not the most important, manufacturing process based on machining in today’s industry. Since this is a continuous process, mechanical tool wear becomes an active concern that affects the final product quality. Tool condition monitoring (TCM) is critical in milling to enhance product quality and prevent tool failures, while tool wear estimation is the first step to achieving a complete system. Several intelligent models have been proposed in the literature to estimate mechanical tool wear, but the use of different combinations of recent algorithms in a hybrid way has not been entirely explored. This study introduces a hybrid tool wear estimation model for milling, combining an Adaptive Neuro-Fuzzy Inference System (ANFIS) with Micro Evolutionary Particle Swarm Optimization (MEPSO) on both antecedent and consequent parameters. MEPSO is a recent modification of the classic particle swarm optimization that has not yet been applied in tool wear estimation or in a hybrid setting with ANFIS. The proposed hybrid ANFIS-MEPSO model addresses tool wear prediction challenges by optimizing ANFIS parameters through MEPSO while employing a feature selection method in the ANFIS-MEPSO inputs. A dataset containing experimental milling data, including vibration, acoustic emission, and electrical current signals, was used to train and test the model with 8 features extracted and a feature selection algorithm to estimate tool wear, achieving promising results with MSE, MAE, and R2 metrics of 0.0045, 0.0578, and 0.9058 on testing data. A sensitivity analysis was conducted to better understand the changes in performance by modifying the optimization hyperparameters. It was revealed that stable and optimal performance occurred with σ and γ between 0.6–0.8 and at least 20 particles, while higher values (0.9) led to inconsistent results. This research highlights ANFIS-MEPSO’s viability for tool wear monitoring in milling, suggesting potential for broader application and further refinement.