Granulate material sorting is a mature and well-developed topic, due to its presence in various fields, such as the recycling, mining, and food industries. However, sorting can be improved, and artificial intelligence has been used for this purpose. This paper presents the development of an efficient sorting system for transparent polycarbonate plastic granulate, based on machine vision and air separation technology. The developed belt-type system is composed of a transparent conveyor with an integrated vision camera to detect defects in passing granulates. The vision system incorporates an industrial camera and backlight illumination. Individual particle localization and classification with the k-Nearest Neighbors algorithm were performed to determine the positions and conditions of each particle. Particles with defects are further separated pneumatically as they fall from the conveyor belt. Furthermore, an experiment was conducted whereby the combined performances of our sorting machine and classification method were evaluated. The results show that the developed system exhibits promising separation capabilities, despite numerous challenges accompanying the transparent granulate material.

In industry, metal workpieces are often heat-treated to improve their mechanical properties, which leads to unwanted deformations and changes in their geometry. Due to their high hardness (60 HRC or more), conventional bending and rolling straightening approaches are not effective, as a failure of the material occurs. The aim of the research was to develop a predictive model that predicts the change in the form of a hardened workpiece as a function of the arbitrary set of strikes that deform the surface plastically. A large-scale laboratory experiment was carried out in which a database of 3063 samples was prepared, based on the controlled application of plastic deformations on the surface of the workpiece and high-resolution capture of the workpiece geometry. The different types of input data, describing, on the one hand, the performed plastic surface deformations on the workpieces, and on the other hand the point cloud of the workpiece geometry, were combined appropriately into a form that is a suitable input for a U-Net convolutional neural network. The U-Net model’s performance was investigated using three statistical indicators. These indicators were: relative absolute error (RAE), root mean squared error (RMSE), and relative squared error (RSE). The results showed that the model had excellent prediction performance, with the mean values of RMSE less than 0.013, RAE less than 0.05, and RSE less than 0.004 on test data. Based on the results, we concluded that the proposed model could be a useful tool for designing an optimal straightening strategy for high-hardness metal workpieces. Our results will open the doors to implementing digital sustainability techniques, since more efficient handling will result in fewer subsequent heat treatments and shorter handling times. An important goal of digital sustainability is to reduce electricity consumption in production, which this approach will certainly do.

This study proposes a novel approach to predict the shape of hardened metal workpieces using acoustic responses processed by a deep convolutional neural network (CNN), aiming to advance automated straightening in manufacturing. Tool steel 1.2379 workpieces of varying widths (24 mm, 90 mm, 200 mm) were struck using a custom-built device, with acoustic responses captured and transformed into scalograms via Continuous Wavelet Transform (CWT). A 40-layer CNN predicted 5 × 9 shape matrices, validated by 3D scans. The dataset (219 shape states, 3396 recordings) was evaluated using leave-one-workpiece-out cross-validation, comparing the CNN against baseline models (linear regression, random forest, shallow CNN, XGBoost). CNN achieved competitive accuracy, demonstrating the feasibility of acoustic-based shape prediction. As a non-invasive, cost-effective complement to 3D scanning, this method offers innovative potential for multi-modal quality control systems in manufacturing.

Potato sorting in post-harvest processing relies heavily on manual visual inspection, which is physically demanding, subjective, and insufficiently scalable for modern packing lines. This study investigates the feasibility of a low-cost RGB-based optical inspection system for automated potato quality detection using deep learning-based object detection. A controlled imaging platform was constructed using commodity hardware, and a dataset of 19,805 manually annotated instances across 1361 images was collected from two geographically distinct farm locations in Slovenia. A systematic benchmark of 25 model configurations spanning five YOLO architecture families—YOLOv8, YOLOv9, YOLOv10, YOLOv11, and YOLO26—was conducted across three practical quality classes (Edible, Feed, Rotten) using a strict cross-location evaluation protocol in which models were trained on one location and tested on a completely unseen second location. All models achieved strong in-distribution performance (F1 ≥ 0.906), but showed considerable variation under cross-location conditions, with external F1 ranging from 0.792 to 0.918. The yolo26_l configuration achieved the best cross-location performance (F1 = 0.918, mAP@0.5:0.95 = 0.816, ΔF1 = 0.029), demonstrating that transferable representations are achievable under a standard supervised training protocol. Per-class analysis identified feed detection as the primary generalization bottleneck. The results confirm that affordable RGB-based sorting systems are technically feasible and highlight cross-location evaluation as an essential protocol for assessing real-world deployment readiness.

Generating numerical control (NC) code from computer-aided design (CAD) models is one of the most knowledge-intensive tasks in digital manufacturing, yet how explicit the input representation should be has received little empirical attention. This study encodes the same machining cases at three levels of increasing explicitness — geometry only (L1), geometry with a tool library (L2), and geometry with both a tool library and an operation-level process plan (L3) — on a deterministic dataset of 9,000 prismatic workpieces, fine-tunes a CodeT5 + 220 M encoder–decoder for each level under five training seeds, and evaluates 900 held-out cases through a hierarchical five-level framework from syntax to geometric correspondence. Explicitness had no robust effect on single-sample correctness: mean geometric-correspondence pass rates were 0.954, 0.952, and 0.969 for L1, L2, and L3, a spread an order of magnitude smaller than the between-seed range, and the L1–L3 contrast was significant in opposite directions under two seeds. Its effect on reproducibility was consistent: the geometric reproducibility of L3 stayed at 0.95 or above under every seed and every decoding temperature up to 0.8, whereas L1 and L2 matched it under four seeds and collapsed to about 0.55 under the fifth. An explicitness index counting the process decisions a representation supplies rather than leaves to be inferred reproduces this pattern quantitatively, and a deterministic rule-based reference shows that the omitted decisions can equally be supplied by process-planning rules. Explicitness therefore acts primarily as a stabiliser of reproducibility rather than as a means of improving single-program correctness.