Search

PURPOSE: The aim of the article is to develop an algorithm for forecasting sales in the supply chain based on the LSTM network using historical sales data of a furniture industry company., DESIGN/METHODOLOGY/APPROACH: Machine learning was used to analyze the data. The method of predicting the behavior of sales value in a specific time horizon in terms of a time series was presented. The LSTM network was used to predict sales. The network used is a special case of recursive neural networks with an important difference in the repeating module. Due to the fact that the activities are carried out on time series, the data was analyzed in terms of the stationarity of such series or trends and seasonal effects. The data used in the analysis includes the daily sales values of a group of certain furniture collections over a specified time horizon. The stationarity of the time series can have a significant impact on its properties and behavior prediction, where failure to bring the time series to the correct form of stationarity can lead to false results., FINDINGS: The result of the research was the analysis of sales forecasting in the supply chain based on machine learning. As a result of the data transformations, the algorithm was able to recognize and learn long-term relationships., PRACTICAL IMPLICATIONS: The presented method of predicting the behavior of sales value in a specific time horizon allows for a look at the forecasting of demand in terms of the supply chain. The sales data of a company from the furniture industry were used for the analysis., ORIGINALITY/VALUE: A novelty is the use of the LSTM network trained on real transaction data of a furniture company that has based its business on the supply chain and cooperates with its suppliers and recipients in Central and Eastern Europe., peer-reviewed

PURPOSE: The aim of the article is to develop a system for analyzing processes and data from various data sources based on machine learning methods., DESIGN/METHODOLOGY/APPROACH: For data analysis, models for forecasting the level of sales and the pipeline gathering operations for the preparation of features, data, model training and its verification were designed. The data analysis pipeline is built of stages related to the preparation of features and data preparation. The learning process requires a specific division of data. The data set has been divided into three subsets, i.e., the training and validation data used in the learning process and the test subset used to verify the quality of the model., FINDINGS: The results of the conducted research show that the use of this type of analytical methods allows for the creation of new business processes, adaptation of services and goods to customer requirements, or the appropriate location of products on the retail space in order to optimize the time of shopping (especially taking into account the pandemic situation)., PRACTICAL IMPLICATIONS: The models presented in the article can be used by combining sales systems and behavioral data related to the movement of customers in the area of the sales space, where it is possible to build systems that allow optimization of orders, the way of arranging goods and other customer behavior patterns., ORIGINALITY/VALUE: A novelty is the construction of a multi-source model for data analysis, where appropriate predictive models were built to predict the level of sales with the use of machine learning algorithms., peer-reviewed

Monitoring of industrial processes is an important element ensuring the proper maintenance of equipment and high level of processes reliability. The presented research concerns the application of the deep learning method in the field of ultrasound tomography (UST). A novel algorithm that uses simultaneously multiple classification convolutional neural networks (CNNs) to generate monochrome 2D images was developed. In order to meet a compromise between the number of the networks and the number of all possible outcomes of a single network, it was proposed to divide the output image into 4-pixel clusters. Therefore, the number of required CNNs has been reduced fourfold and there are 16 distinct outcomes from single network. The new algorithm was first verified using simulation data and then tested on real data. The accuracy of image reconstruction exceeded 95%. The results obtained by using the new CNN clustered algorithm were compared with five popular machine learning algorithms: shallow Artificial Neural Network, Linear Support Vector Machine, Classification Tree, Medium k-Nearest Neighbor classification and Naive Bayes. Based on this comparison, it was found that the newly developed method of multiple convolutional neural networks (MCNN) generates the highest quality images.

The work presents image reconstruction in ultrasonic transmission tomography using the Block-Wise-Transform-Reduction method. The system consists of a tomography platform using ultrasound built by the authors, and an algorithm that can reconstruct images in a distributed system. The algorithm applies image compression techniques (Discrete Cosine Transformation) for each block of the image separately. This allows location object in the analyzed area for real-time image reconstruction using an ultrasound tomographic device. The idea behind the method used is that the reconstruction process is directly connected to the compression process.

This paper refers to the method of using the deep neural long-short-term memory (LSTM) network for the problem of electrocardiogram (ECG) signal classification. ECG signals contain a lot of subtle information analyzed by doctors to determine the type of heart dysfunction. Due to the large number of signal features that are difficult to identify, raw ECG data is usually not suitable for use in machine learning. The article presents how to transform individual ECG time series into spectral images for which two characteristics are determined, which are instantaneous frequency and spectral entropy. Feature extraction consists of converting the ECG signal into a series of spectral images using short-term Fourier transformation. Then the images were converted using Fourier transform again to two signals, which includes instantaneous frequency and spectral entropy. The data set transformed in this way was used to train the LSTM network. During the experiments, the LSTM networks were trained for both raw and spectrally transformed data. Then, the LSTM networks trained in this way were compared with each other. The obtained results prove that the transformation of input signals into images can be an effective method of improving the quality of classifiers based on deep learning.

Akkermansia muciniphila, an abundant member of the human gut microbiota, has been suggested as a potential next-generation probiotic. However, its high sensitivity to oxygen limits the development of dosage protocols. Here, we describe microencapsulation, in a xanthan and gellan gum matrix, and a subsequent freeze-drying protocol for A. muciniphila DSM22959. For comparison Lactobacillus plantarum subsp. plantarum ATCC14917 was microencapsulated and freeze-dried using similar protocols. Four different mixtures were tested for cryoprotective properties: sucrose 5% plus trehalose 5%; agave syrup 10%; skim milk 10%, glucose 1%, yeast extract 0.5%, and mannitol 2.5%; as well as peptone 0.1% plus sorbitol 1.2%. Milli-Q-water served as control. Only cryoprotectant solutions with high sugar or protein content significantly improved the survival of both strains during freeze-drying. Microencapsulated cells were stored aerobically or anaerobically for 1 month at 4 °C or 25 °C. Survival of A. muciniphila was significantly better when stored anaerobically at 4 °C. The survival of microencapsulated L. plantarum, was relatively stable at both temperatures under anaerobic conditions. Survival of microencapsulated cells was compared with that of free cells during in vitro simulated upper gastrointestinal tract (GIT) transit at fasted and fed state. During in vitro simulated stomach passage, encapsulation significantly improved survival and viable cells remained at relevant levels after the entire simulated upper GIT transit. In conclusion, we here report a protocol for encapsulating A. muciniphila giving acceptable storage stability and enhancing survival during in vitro simulated upper GIT transit and thus constitutes an important step towards enabling future use of this important member of the human colonic microbiota as a probiotic.

Simple Summary Weaning is the most stressful event in pig production and is often associated with reduced performance, diarrhoea and piglet mortality. Currently, a high dose of zinc oxide (ZnO) is used to prevent weaning-related loss in productivity. However, the feeding of ZnO in weaner piglets will be phased out by 2022 in Europe, leaving pig producers without options to manage post-weaning disorders. This study investigated whether fermented rapeseed meal (FRM) alone or in combination with one (FRMA) or more (FRMAS) brown macroalgae species could improve weaner piglet growth, intestinal development and health compared to either non-supplemented diets (negative control, NC) or diets supplemented with 2500 ppm ZnO (positive control, PC). Both FRM and FRMA resulted in a similar production performance to PC when fed to weaned piglets. The PC, FRM and FRMAS (gender-specific) improved jejunal villus development more than the NC. Colon mucosal development was stimulated, and signs of intestinal inflammation were reduced by FRM. The composition and diversity of colon microbiota were similar between all fermented feeds and PC, but different compared to NC. In conclusion, FRM was at least as effective as ZnO to improve piglet growth, intestinal development and health. Abstract The feeding of medicinal zinc oxide (ZnO) to weaner piglets will be phased out after 2022 in Europe, leaving pig producers without options to manage post-weaning disorders. This study assessed whether rapeseed meal, fermented alone (FRM) or co-fermented with a single (Ascophylum nodosum; FRMA), or two (A. nodossum and Saccharina latissima; FRMAS) brown macroalagae species, could improve weaner piglet performance and stimulate intestinal development as well as maturation of gut microbiota in the absence of in-feed zinc. Weaned piglets (n = 1240) were fed, during 28–85 days of age, a basal diet with no additives (negative control; NC), 2500 ppm in-feed ZnO (positive control; PC), FRM, FRMA or FRMAS. Piglets fed FRM and FRMA had a similar or numerically improved, respectively, production performance compared to PC piglets. Jejunal villus development was stimulated over NC in PC, FRM and FRMAS (gender-specific). FRM enhanced colon mucosal development and reduced signs of intestinal inflammation. All fermented feeds and PC induced similar changes in the composition and diversity of colon microbiota compared to NC. In conclusion, piglet performance, intestinal development and health indicators were sustained or numerically improved when in-feed zinc was replaced by FRM.

Over the years, the issue of air pollution is increasingly published in line with the seriousness of this problem for human health. However, air quality is just one of many issues that are solved by Environmental Computer Science. The areas in which Environmental Computer Science deals are the foundation for the creation of Environmental Decision Support Systems. Such systems are complex, extensive and multidisciplinary IT solutions. The aim of the research was to develop a comprehensive system based on the Internet of Things architecture, which is used to monitor environmental conditions. The first stage of this research includes the creation of a system for collecting, storing and analyzing data in the area of air quality tests in the city.

The article presents a novel approach to ultrasound tomography in industrial applications. In order to visualize the interior of a tank (reactor) filled with tap water, a single neural network enhanced with autoencoder was used. A novelty is the use of an autoencoder to improve the quality of the measurement vector. Thanks to the use of the autoencoder for denoising the input measurements in connection with the appropriately adapted neural network, the quality of the output image was improved. A robust algorithm was developed that properly reconstructs hidden objects in monochrome images with high efficiency.

This article presents a method of using the logistic regression to solve the inverse problem in electrical impedance tomography. Multicollinearity between predictors in linear models make it difficult to precisely forecast for output variable. To reduce the problem of highest correlation between measures obtained from electrodes the elasticnet method was applied.

In the industrial production of yoghurt, measurement of pH is normally the only in-line technique applied as a real-time monitoring signal for following the dynamics during the fermentation process. However, every dairy company would benefit from an in-line technique giving information about the chemical composition, physical/textural properties and/or microbial contamination. In this study lactic fermentation batches with the starter bacteria Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus are explored by in-line near infrared (NIR) spectroscopy. The dynamics obtained from NIR spectroscopy is elucidated in order to explain what causes the variation of the indirect measurements. The results show that viscosity change has a large impact on the scatter, which affects the NIR data. It seems that the variation found by NIR spectroscopy is caused by both scatter and absorption, where the scatter gives us information about the textural change happening, and the absorption gives us information about the biomass formation plus the conversion of sugar into lactic acid.

In this paper there were presented some information about marketing and advertisement in the aspect of acquiring customers-candidate for studies with the help of the film as a tool of this kind of action. The paper also shows the results of the survey research and their analysis in the scope of expected features which will be characteristic, according to the respondents, for above-mentioned transfer tool.

This paper presents a concept of the air quality monitoring system design and describes a selection of data quality analysis methods. A high level of industrialisation affects the risk of natural disasters related to environmental pollution such ase.g.air pollution by gases and clouds of dust (carbon monoxide, sulphur oxides, nitrogen oxides). That is why researches related to the monitoring this type of phenomena are extremely important. Low-cost air quality sensors are more commonly used to monitor air parameters in urban areas. These types of sensors are used to obtain an image of the spatiotemporal variability in the concentration of air pollutants. Aside from their low price , which is important from a point of view of the economic accessibility of society, low-cost sensors are prone to produce erroneous results compared to professional air quality monitors. The described study focuses on the analysis of outliers as particularly interesting for further analysis, as well as modelling with machine learning methods for air quality assessment in the city of Lublin.

There is a growing interest in understanding the fate and behaviour of probiotic microorganisms and bioactive compounds during passage of the human gastrointestinal tract (GIT). Here, we report the development of a small volume in vitro model called The smallest Intestine (TSI) with increased throughput focusing on simulating passage through the stomach and small intestine (SI). The basic TSI module consists of five reactors, with a working volume of 12 ml each. During the simulated passage through the SI, bile is absorbed and pH is adjusted to physiologically relevant values for duodenum, jejunum and ileum. A consortium of seven representative bacterial members of the ileum microbiota is included in the ileal stage of the model. The behaviour of three putative probiotic Lactobacillus strains during in vitro simulated upper GIT passage was tested in the model and results were compared to previous studies describing probiotic survival. It was found, that probiotic persistence is strongly related to whether food was ingested, but also to presence of the ileal microbiota, which significantly impacted probiotic survival. In conclusion, TSI allows testing a substantial number of samples, at low cost and short time, and is thus suitable as an in vitro screening platform.

The article shows a case study of Cloud Computing model combined with AT-Mega microcontrollers for educational purposes. The presented system takes advantage of many aspects of Internet of Things model, thus conjoining Cloud Management system with measurement-execution module based on Arduino platform. One benefit of this solution is a cost-effective way of showcasing machine and device integration with distinct cloud services. This article is based on practical experience with students' projects and an home alarm system with use of a Cloud Computing services will be described.

This paper presents generic concept of using measuring devices of using cloud-based services as a basic application of the Internet of Things (IoT) model. This type of applications will become a base of future of the Web of Things paradigm. A high level of operational infrastructure is outlined and corresponding technologies are illustrated. To demonstrate the data provision, a cloud-based solution is described along with details of its design and operation. A number of technology, including communication protocols, transportation protocols and middleware are depicted as a basic introduction of implementation of future WoT solutions.