Automatic Production Research Articles

The Development Potential and Challenges of One-Click Generative Artificial Intelligence in Cross-Border E-Commerce

This paper explores the development potential and challenges of one-click generative artificial intelligence (AI) in cross-border e-commerce. With the rapid growth of the Internet, cross-border e-commerce has seen significant market expansion, and the integration of AI technology presents new opportunities for the industry. This study analyzes the potential of one-click generative AI in cross-border e-commerce and its challenges. Reviewing relevant academic literature, the paper outlines key application scenarios of one-click generative AI, such as automatic product description generation and intelligent customer service assistants. Additionally, the study highlights the challenges associated with the technology, including issues related to semantic understanding and technological immaturity. The findings indicate that while one-click generative AI holds great potential for cross-border e-commerce, both technical and legal challenges must be addressed for its successful implementation.

PRODUCTIVE ROBOTS AND INDUSTRIAL EMPLOYMENT: THE ROLE OF NATIONAL INNOVATION SYSTEMS

AbstractIn a model with robots, automatable and nonautomatable production, we study robot‐labor substitutions and show how they are influenced by a country's “innovation system.” Substitution depends on demand and production elasticities, the country's innovation capabilities, and openness. Making use of World Economic Forum data, we estimate the relationship for 13 countries and find that countries with poor innovation capabilities substitute robots for workers much more than countries with richer innovation capabilities, which might complement them. Innovation capabilities play a bigger role in the high‐tech electronics sector than in other manufacturing and play a limited role in nonmanufacturing.

Automatic batching compound fertilizer production system based on parallel bin structure

Automatic batching compound fertilizer production system based on parallel bin structure

Ampycloud: an open-source algorithm to determine cloud base heights and sky coverage fractions from ceilometer data

Abstract. Ceilometers are used routinely at aerodromes worldwide to derive the height and sky coverage fraction of cloud layers. This information, possibly combined with direct observations by human observers, contributes to the production of meteorological aerodrome reports (METARs). Here, we present ampycloud, a new algorithm, and its associated Python package for automatic processing of ceilometer data with the aim of determining the sky coverage fraction and base height of cloud layers above aerodromes. The ampycloud algorithm was developed at the Swiss Federal Office of Meteorology and Climatology (MeteoSwiss) as part of the AMAROC (AutoMETAR/AutoReport rOund the Clock) program to help in the fully automatic production of METARs at Swiss civil aerodromes. ampycloud is designed to work with no direct human supervision. The algorithm consists of three distinct, sequential steps that rely on agglomerative clustering methods and Gaussian mixture models to identify distinct cloud layers from individual cloud base hits reported by ceilometers. The robustness of the ampycloud algorithm stems from the first processing step, which is simple and reliable. It constrains the two subsequent processing steps that are more sensitive but also better suited to handling complex cloud distributions. The software implementation of the ampycloud algorithm takes the form of an eponymous, pip-installable Python package developed on GitHub and made publicly accessible.

Obtaining estimation algorithms for water quality variables in the Jaguari-Jacareí Reservoir using Sentinel-2 images

Satellite images are essential tools for monitoring aquatic ecosystems and assessing water quality, as they enable the measurement of parameters such as chlorophyll-a (Chl-a) concentration, phycocyanin (PC), and cyanobacteria density. These indicators aid in evaluating eutrophication processes and detecting cyanobacteria in aquatic ecosystems. This study utilized field data and images captured by the Sentinel-2 sensor from 2015 to 2022 to investigate the Jaguari-Jacareí reservoirs (JAG-JAC). Two atmospheric corrections from the Case 2 Regional Coast Color (C2RCC) processor, namely C2X and C2XC, were applied, and algorithms were developed to estimate the parameters using both in situ data measurements and reflectance data extracted from the images. For Chl-a concentration, the dataset was divided into two blocks: one for model calibration (70% of the data) and the other for validation (30% of the data). As for PC, the entire dataset was utilized to calibrate the model, and validation was conducted through cross-validation using the Automated Radiative Transfer Model Operator (ARTMO) software. Cyanobacteria density was indirectly estimated from the Chl-a concentrations determined in the field samples, as these variables exhibited a strong correlation, also validating the model previously proposed for the Cantareira system for estimating cyanobacteria density from Chl-a data. Additionally, the automatic chlorophyll-a products (con_chla) derived from the C2X and C2XC processors were validated. The findings revealed that the C2X processor exhibited the greatest potential for estimating water quality parameters. It was observed that the most effective algorithms were derived using the R705/R665 band ratio for Chl-a and the R705/R490 ratio for PC. For cyanobacteria density, the optimal algorithm was established based on the relationship between cyanobacteria density and Chl-a using the data obtained in this study.

Coupled optimization of task sequence and hoist scheduling for electroplating production lines based on an improved salp swarm algorithm

Automatic electroplating production lines have been widely used in electronics industries to reduce the labour intensity and improve the production efficiency. In the multi-variety and low-volume electroplating production, it is known that the task loading sequence and hoist scheduling are coupled with each other, and they codetermine the production efficiency, while all the existing scheduling methods consider them separately, and thus the optimal production schemes become unavailable. Therefore, this paper develops a Task sequence-Hoist scheduling Coupled Optimization (THCO) model which simultaneously considers the requirements and practical constrains of task sequence and hoist scheduling, having an optimization objective of minimizing the maximum completion time. For this model, a double-layer code is developed and an Improved Salp Swarm Algorithm (ISSA) is developed by introducing three improvement strategies: the random spare strategy which is used to increase the population diversity, the nonlinear adaptive weight strategy which is used to balance the exploration and exploitation capacities, and a golden sine algorithm which is used to improve the convergence rate. Experiments based on 23 benchmark functions are then conducted. The obtained results show that ISSA has better convergence and solving quality than existing algorithms. Furthermore, several production cases prove that THCO can generate production schemes that better meet the requirements of production lines.

Research on Forging Dimension Online Measuring System Based on Vibration Point Cloud Compensation

Mechanical vibration in the high-temperature forging production line often causes large forging thermal dimensional measurement error in the detection task, so a vibration point cloud compensation method based on an acceleration sensor is proposed in this study. First, the vibration signal is obtained through the built-in acceleration sensor in the laser camera. After the acceleration of the camera vibration is detected, the displacement of the camera in three directions is solved by secondary integration. Subsequently, the coordinate value of the corresponding point is obtained by the rotation matrix transformation so as to compensate and correct the point cloud deviation caused by the camera vibration. Finally, the forging point cloud is matched using the surface matching algorithm in Halcon. An automatic forging production line for wheel hubs has been built, and the key dimensions of high-temperature forging products have been measured online using the developed method. After the forging point cloud is compensated, the average measurement error of dimensions is reduced from ±0.9 mm to ±0.1 mm, and the standard deviation is reduced from 0.52 mm to 0.056 mm. Using the vibration point cloud compensation method based on the acceleration sensor, as well as using silica aerogel insulation, vibration structural parts, heat insulation and constant temperature, a blue-violet 3D laser camera, and other measures, the dimensional detection accuracy of high-temperature forgings in the forging production line can be improved, and the instability of dimensional detection can be reduced.

Landslide Susceptibility Assessment by Machine Learning and Frequency Ratio Methods Using XRAIN Radar-Acquired Rainfall Data

This study is an efficiency comparison between four methods for the production of landslide susceptibility maps (LSMs), which include random forest (RF), artificial neural network (ANN), and logistic regression (LR) as the machine learning (ML) techniques and frequency ratio (FR) as a statistical method. The study area is located in the Southern Hiroshima Prefecture in western Japan, a locality known to suffer from rainfall-induced landslide disasters, the most recent one in July 2018. The landslide conditioning factors (LCFs) considered in this study are lithology, land use, altitude, slope angle, slope aspect, distance to drainage, distance to lineament, soil class, and mean annual precipitation. The rainfall LCF data comprise XRAIN (eXtended RAdar Information Network) radar records, which are novel in the task of LSM production. The accuracy of the produced LSMs was calculated with the area under the receiver operating characteristic curve (AUROC), and an automatic hyperparameter tuning and result comparison system based on AUROC scores was utilized. The calculated AUROC scores of the resulting LSMs were 0.952 for the RF method, 0.9247 for the ANN method, 0.9016 for the LR method, and 0.8424 for the FR. It is also noteworthy that the ML methods are substantially swifter and more practical than the FR method and allow for multiple and automatic experimentations with different hyperparameter settings, providing fine and accurate outcomes with the given data. The results evidence that ML techniques are more efficient when dealing with hazard assessment problems such as the one exemplified in this study. Although the conclusion that the RF method is the most accurate for LSM production as found by other authors in the literature, ML method efficiency may vary depending on the specific study area, and thus the use of an automatic multi-method LSM production system with hyperparameter tuning such as the one utilized in this study is advised. It was also found that XRAIN radar-acquired mean annual precipitation data are effective when used as an LCF in LSM production.

StereoThermoLegs: label propagation with multimodal stereo cameras for automated annotation of posterior legs during running at different velocities

In sports science, thermal imaging is applied to investigate various questions related to exercise-induced stress response, muscle fatigue, anomalies, and diseases. Infrared thermography monitors thermal radiation from the skin’s surface over time. For further analysis, regions of interest are extracted and statistically analyzed. Although computer vision algorithms have grown in recent years due to data-driven approaches, this is not the case for detailed segmentation in thermal images. In a supervised manner, machine learning optimizations require a large amount of training data with input and ground truth output data. Unfortunately, obtaining annotated data are a costly problem that increases with the complexity of the task. For semantic segmentation, pixel-wise label masks must be created by experts. Few datasets meet the needs of sports scientists and physicians to perform advanced applications of thermal computer vision during physical activity and generate new insights in their fields. In this paper, a new method is introduced to transfer segmentation masks from the vision domain to the thermal domain with a stereo-calibrated time-of-flight camera and high-resolution mid-wave infrared camera. A post-processing procedure is then utilized to obtain dense pixel masks for the posterior legs during walking and running on a treadmill. The developed StereoThermoLegs dataset is based on 14 participants and includes 11 subjects for training with 12,826 thermograms and the remaining three individuals for testing with 3433 images. A deep neural network was trained with the DeepLabv3+ architecture, the AdaBelief optimizer, and Dice loss as a benchmark. After 29 epochs, the test set achieved an average intersection over union of 0.66. The analysis of the posterior leg region, specifically the left and right calf, offered the most insights, with values of 0.83 and 0.83, respectively. The first multimodal stereo dataset containing synchronized visual and thermal images of a runner’s back provides a starting point for data-driven segmentation tasks in sports science and medicine. Our technique allows for automatic production of customized datasets for deep learning, accelerating the implementation of baseline outcomes for newly identified areas of interest in thermal imaging, while bypassing the requirement for extensive manual annotation. The approach is not exclusive to stereo rig and segmentation tasks utilizing RGBD and thermal cameras, but can be applied to other imaging tasks and modalities.

165 Ergonomic intervention for maintenance workers for an automatic production system: a case study

165 Ergonomic intervention for maintenance workers for an automatic production system: a case study

A triboelectric sensing array integrating material identification and self-healing enabled by a healable polyamide-based device unit

To develop skin-like tactile sensor systems with both self-healing ability of repairing mechanical damage and perception ability of identifying materials and objects is significantly valuable for smart and high-quality interaction with environment. It is noticed that the sensing array comprised of distinct triboelectric nanogenerator (TENG) units shows great advantages in material recognition. However, those attempts of integrating self-healing ability into above triboelectric sensing array are challenged by the lack of self-healing TENG constituent units with differentiated electron affinity. Herein, we exploit a self-healing polyamide (SPA) with high mechanical strength as the base material for constructing a novel self-healing TENG unit with distinctive electron affinity. The healing of the prepared SPA-based TENG (SPA-TENG) can be autonomously conducted at room temperature without external intervention required. Our results reveal that SPA-TENG possesses stronger electron affinity compared to existing self-healing TENGs, which enables it to be an ideal constituent unit to construct self-healing triboelectric sensing array for material identification. After combining SPA-TENG with other previously reported self-healing TENGs, we firstly demonstrate a fully self-healing triboelectric sensing array capable of distinguishing various commercial plastic products based on their fingerprint waveform combinations, which shows promising application in fields of intelligent robots, artificial prosthesis, and automatic production etc.

Datasets and methods of product recognition on grocery shelf images using computer vision and machine learning approaches: An exhaustive literature review

A product recognition system recognizes all products on the shelf images and determines their positions. A business equipped with an automatic product recognition system has a convenient follow-up of many human-powered activities while increasing customer satisfaction. That is, product recognition stands out with its benefits such as tracking shelf layouts and stocking their status, and improving the shopping experience for customers, especially the visually impaired ones. However, product recognition is a challenging problem of computer vision in terms of the difficulty of obtaining and updating datasets and the breadth of the product scale. On the other hand, the number of studies on product recognition is constantly increasing by using various computer vision and machine learning methods, and effective solutions are offered to this problem. This paper provide a comprehensive review in the field of researches of product recognition on grocery store shelves. In this article, data sets and approaches used in the literature for the development of an automatic product recognition system are examined and compared, and their benefits and limitations are commented. Finally, a guideline is provided for future researchers and new perspectives for future studies are presented.

A novel transformer‐based graph generation model for vectorized road design

AbstractRoad network design, as an important part of landscape modeling, shows a great significance in automatic driving, video game development, and disaster simulation. To date, this task remains labor‐intensive, tedious and time‐consuming. Many improved techniques have been proposed during the last two decades. Nevertheless, most of the state‐of‐the‐art methods still encounter problems of intuitiveness, usefulness and/or interactivity. As a rapid deviation from the conventional road design, this paper advocates an improved road modeling framework for automatic and interactive road production driven by geographical maps (including elevation, water, vegetation maps). Our method integrates the capability of flexible image generation models with powerful transformer architecture to afford a vectorized road network. We firstly construct a dataset that includes road graphs, density map and their corresponding geographical maps. Secondly, we develop a density map generation network based on image translation model with an attention mechanism to predict a road density map. The usage of density map facilitates faster convergence and better performance, which also serves as the input for road graph generation. Thirdly, we employ the transformer architecture to evolve density maps to road graphs. Our comprehensive experimental results have verified the efficiency, robustness and applicability of our newly‐proposed framework for road design.

Rancang Bangun Pengemasan Produk Kacang Goreng Otomatis Berbasis Mikrokontroler

The working system of this Automatic Fried Bean Product Packaging tool uses Arduino Mega2560 as a microcontroller, then a screw on the funnel to accelerate the descent of the product into the packaging plastic and prevent blockages, infrared sensors to detect product availability in storage containers. The tool made uses volume measurements to minimize error on the scales using a load cell sensors. The type of research used in making the final assignment is action research which is carried out carefully, logically and systematically. The conclusions of this research are (1) The system can detect the supply of fried peanuts in the funnel/silo using infrared FC-51 sensor with a distance range of 1 – 8 cm. (2) The system can read the mass input selected by user, namely 75 g, 100 g, and 150 g with a push button and displayed through the LCD. (3) The system can fill the fried peanuts into packages according to the mass that has been set through the KY-008 laser sensor and laser receiver, using the tube volume formula with a percentage error of 4.5% for a mass of 75 grams, 3.83% for a mass of 100 grams, and 5.41% for a mass of 150 grams. (4) The system can perform the package closing process using servo motors, heating elements, and stepper motors. The duration of time required by the heating element in gluing the plastic ranges from 3 to 4 seconds.

НАПРАВЛЕНИЯ РАЦИОНАЛЬНОЙ ЛОКАЛИЗАЦИИ ОБОРУДОВАНИЯ ДЛЯ ПРОИЗВОДСТВА ПЕЛЬМЕНЕЙ

The scientific article contains a theoretical analysis of scientific literature on the problems of ensuring the economic efficiency of automation and robotization of technological processes in the industrial production sector, their impact on the socio-economic situation and the economic system as a whole. Directions for the effective organization of dumpling production are presented, the most important of which is the rational localization of equipment. Attention is drawn to the need to invest in the process of robotization or automation of the production of dumplings, and robotization is several times more expensive than conventional automatic equipment. The technological processes for the production of dumplings are considered and equipment capable of robotizing or automating this process is shown. At the technological stage of meat chopping, the importance of using universal equipment for chopping meat and vegetables such as a cutter is shown. The dough preparation step is shown using a robotic system capable of preparing the dough or starter and then the dough, equipped with a robotic arm for mixing the bowl. For comparison, a more budget-friendly option is presented - a traditional automatic dough kneading machine. Options for localizing equipment for preparing dumplings are considered: robotic equipment for the production of dumplings developed by the Japanese Institute of Technology AICHI and an automatic dumpling production machine that does not require a dough sheeter. It is indicated that the final technological process is shock freezing, for which a special cabinet is used. It was noted that after shock freezing the product is ready for sale, but it is more efficient to package and store it in a freezer before transferring it to the retail chain. The effectiveness of investments is confirmed by the payback period, the calculation of which takes into account the volume of investments and the income expected to be received from the use of the equipment.

An Improved SSD Model for Small Size Work-pieces Recognition in Automatic Production Line

<p>Aiming at the problems of slow recognition speed and low recognition accuracy of arbitrarily placed workpiece by machine vision in traditional automated production lines, a workpiece recognition algorithm based on improved SSD is proposed. Firstly, the improved DarkNet53 is used to replace the backbone network in the original SSD network framework, and the network enhancement is used in the backbone network to solve the defect of small target missed detection. Then, channel attention module and deep semantic feature fusion module are added, in order to improve the recognition ability and detection accuracy of the small target features. Lastly, the loss function was optimized, and the problem caused by sample imbalance was solved by changing the weight distribution of positive and negative samples. In the experiment, image datasets of typical bolts, nuts, and connecting plates were constructed for the network training, the experimental results showed that, the recognition accuracy and speed have been optimized and meet the requirements of automatic work-piece detection in actual production, compared with traditional YOLOv4 and the original SSD algorithm in the work-piece recognition task.</p> <p> </p>

All-Automated Fabrication of Freestanding and Scalable Photo-Thermoelectric Devices with High Performance.

Flexible photo-thermoelectric (PTE) devices have great application prospects in the fields of solar energy conversion, ultrabroadband light detection, etc. A suitable manufacturing process to avoid the substrate effects as well as to create a narrow transition area between p-n modules for high-performance freestanding flexible PTE devices is highly desired. Herein, an automated laser fabrication (ALF) method is reported to construct the PTE devices with rylene-diimide-doped n-type single-walled carbon nanotube (SWCNT) films. Thewet-compressing approach is developed to improve the thermoelectric power factors and figure of merit (ZT) of the SWCNT hybrid films. Then, the films are cut and patterned automatically to make PTE devices with various structures by the proposed ALF method. The freestanding PTE device with a narrow transition area of ≈2-3 µm between the p and n modules exhibits a high-power density of 0.32 µW cm-2 under the light of 200 mW cm-2, which is among the highest level for freestanding-film-based PTE devices. The results pave the way for the automatic production process of PTE devices for green power generation and ultrabroadband light detection.

Improvement of lithium battery corner detection accuracy based on image restoration method

Target detection technology has been widely used in the automatic production of lithium batteries. However, motion blur will lead to the reduction of the angular position detection accuracy of lithium batteries. To solve this problem, an improved fuzzy recovery model for angular position of lithium battery is proposed in this paper. Firstly, the improved lightweight neural network RepVGG was used as the main module of the backbone network, so that the network could improve the performance of network feature extraction while reducing the number of calculation parameters and improving the reasoning speed of fuzzy restoration. Secondly, we optimize the multi-Dconv head transposed attention (MDTA) module and reference it to the generator, which reduces the complexity of the model and strengthens the network’s attention to details and textures, and improves the visual effect of the restored image. Finally, we design a lightweight globally connectable residual network called SAC Block and use it to to improve the discriminator, which enhances the global receptive field of the model and improves the structural similarity between the restored image and the original image. In order to verify the effectiveness of the method, we verify it on the self-built dataset and GoPro dataset. The experiments show that our proposed lightweight model improves the peak signal-to-noise ratio (PSNR) index by 9.2% and 8.6% respectively compared with the original model. The visual effect of the restored image is better than that of other current similar algorithms, and it is confirmed that our model can better improve the accuracy of lithium battery angular position detection.

A hyperspectral deep learning attention model for predicting lettuce chlorophyll content

BackgroundThe phenotypic traits of leaves are the direct reflection of the agronomic traits in the growth process of leafy vegetables, which plays a vital role in the selection of high-quality leafy vegetable varieties. The current image-based phenotypic traits extraction research mainly focuses on the morphological and structural traits of plants or leaves, and there are few studies on the phenotypes of physiological traits of leaves. The current research has developed a deep learning model aimed at predicting the total chlorophyll of greenhouse lettuce directly from the full spectrum of hyperspectral images.ResultsA CNN-based one-dimensional deep learning model with spectral attention module was utilized for the estimate of the total chlorophyll of greenhouse lettuce from the full spectrum of hyperspectral images. Experimental results demonstrate that the deep neural network with spectral attention module outperformed the existing standard approaches, including partial least squares regression (PLSR) and random forest (RF), with an average R2 of 0.746 and an average RMSE of 2.018.ConclusionsThis study unveils the capability of leveraging deep attention networks and hyperspectral imaging for estimating lettuce chlorophyll levels. This approach offers a convenient, non-destructive, and effective estimation method for the automatic monitoring and production management of leafy vegetables.

Design of an Intelligent Feeding System

The article elaborates on the design method of an intelligent feeding system . The feeding system is an important part of the automatic production line, which provides raw materials for other units of the automatic production line. It is controlled by PLC, which is described from many aspects, such as structure design, pneumatic circuit design, control circuit design and PLC program design. It analyzes the design idea and principle of intelligent feeding system in detail, and provides an effective reference for the design of automatic production line.