Production Forecasts Research Articles

Analysing Forecasts of Sustainable Agri-Food Consumption, Production and Hunger

Analysing Forecasts of Sustainable Agri-Food Consumption, Production and Hunger

КОНКУРЕНТОСПРОМОЖНІСТЬ ПІДПРИЄМСТВА ТА ФАКТОРИ, ЩО ЇЇ ФОРМУЮТЬ

The article investigates and substantiates the theoretical aspects of the essence of competitiveness and the study of internal and external environment factors that affect the competitiveness of an enterprise in market conditions. On the basis of theoretical studies, the author summarizes some of the author's definitions of the concept of “enterprise competitiveness”. It has been determined that the competitiveness of enterprise is characterized by its ability and efficiency to adapt to the conditions of the competitive environment. The features that characterize the competitiveness of an enterprise include: comparability, spatiality, dynamism, objectivity, attributability, systematicity, objectivity. Observance of these characteristics is a prerequisite for conducting analytical research, a key to solving problems and achieving the company's goals. It is proposed to use the modified classification of factors of competitiveness of a firm proposed by O. Kuzmin, which provides for the allocation of the following criterion (integral) groups of factors: environment of the firm's functioning; substantive characteristics; level of regulation of factors. The environmental factors are reasonably divided into direct factors that have a direct impact on competitiveness and indirect factors that have an indirect impact through direct factors. An analysis of the impact of external and internal factors on the company's competitiveness helps to reduce uncertainty and risks in the company's production and sales process, improve the quality of strategic planning and forecasting of the company's production and sales activities, increase the competitiveness of the company and its products, maintain the company's competitive position and increase its market share. The importance of certain substantive factors of competitiveness may vary depending on a number of objective factors, such as the purchasing power of the population, changes in regulation and taxation, market conditions, etc.

System Identification Techniques for Soft Sensors and Multiphase Flow Metering

Production rates estimation and forecasting is a complex problem in multiphase flow systems, such as oil wells. In the oil and gas industry, this task is normally performed by first principles simulation models, which are costly to maintain and adjust. An alternative to this process could be black-box models based on historical data to perform this task. This work proposes to perform an extensive evaluation of existing regressors to create a black-box system identification model to predict oil well liquid flow rates. Improvement strategies such as the creation of error models, time series decomposition, and the use of current-time data are also evaluated. The final results demonstrated the importance of pressure and temperature data for the well flow rate determination and identified the use of current-time data as the best tested technique to improve the performance of the final model.

ARTIFICIAL INTELLIGENCE IN HRM PROCESSES: GENERAL FRAMEWORK AND APPLICATION FOR PSYCHOLOGICAL CLIMATE MONITORING

The article considers the potential for the widespread use of artificial intelligence (AI) to address current challenges in the labor market and improve the efficiency of human resource management (HRM). It analyzes modern approaches to integrating relevant methods and tools into HRM practices. The study explores how AI can complement existing HR processes, creating a more data-driven and adaptive framework for decision-making. Companies are actively seeking ways to incorporate innovative solutions, so the results of this study can be seen as a step towards combining current HRM practices with AI's transformative potential to support effective management decisions. The paper presents powerful AI capabilities: mechanical AI, thinking AI, and feeling AI. Mechanical AI is best suited for automating tasks such as job creation, enrolment in training courses, reporting, monitoring career development, and performance evaluation. This type of AI significantly reduces administrative burdens, allowing HR professionals to focus on strategic decision-making and employee engagement. Thinking AI excels in analyzing CVs, learning outcomes, productivity, and career forecasting. Feeling AI focuses on analyzing applicant behavior, personalizing training programs, providing psychological support, and monitoring the workplace's psychological climate. To address challenges in emotional analysis, the article presents a conceptual model for sentiment analysis of data on the emotional state and satisfaction of employees. The paper includes specific recommendations for implementation and demonstrates the model's application on Glassdoor Job Reviews data using Python programming language. This approach illustrates how AI can enhance employee well-being while aligning HR practices with organizational goals.

Реализация программы социальной газификации и догазификации в преодолении проблем отечественного ТЭК

The material of the proposed article seems to the authors to be extremely relevant in connection with the 11 packages adopted by the Western community against Russia, containing more than 12 thousand of all kinds of sanctions and restrictions, including in relation to the export of hydrocarbons. This circumstance has put the domestic economy in an extremely difficult situation, especially in the context of a special military operation. Reorientation of the oil and gas industry to the East, to the countries of Greater Eurasia, year-round transport and industrial development of the Northern Sea Route and the Arctic zone of Russia will not only overcome the problems and difficulties that have arisen, but also strengthen political and economic ties with the states of the Silk Road and the Organization of the Collective Security Treaty (CSTO). The purpose of the study is to provide a comprehensive economic assessment of the implementation of the state Program for Social Gasification/Additional Gasification in the constituent entities of the Federation, as well as to forecast industrial production based on the processing of hydrocarbon raw materials at different degrees of technological conversion into lines of consumer goods and products intended for export. The research results obtained so far by the authors of the article boil down to a review of information sources on the identified problems, an analysis of individual issues of the dynamics of production, export supplies, transportation and pricing of domestic hydrocarbon raw materials, and to determining the success of the implementation of the gasification program in individual regions of the Federation. The conceptual approaches involved were discussed. In conclusion, the article is accompanied by a number of author's conclusions.

Deep Learning Parameterization of Vertical Wind Velocity Variability via Constrained Adversarial Training

Abstract Atmospheric models with typical resolution in the tenths of kilometers cannot resolve the dynamics of air parcel ascent, which varies on scales ranging from tens to hundreds of meters. Small-scale wind fluctuations are thus characterized by a subgrid distribution of vertical wind velocity W with standard deviation σW. The parameterization of σW is fundamental to the representation of aerosol–cloud interactions, yet it is poorly constrained. Using a novel deep learning technique, this work develops a new parameterization for σW merging data from global storm-resolving model simulations, high-frequency retrievals of W, and climate reanalysis products. The parameterization reproduces the observed statistics of σW and leverages learned physical relations from the model simulations to guide extrapolation beyond the observed domain. Incorporating observational data during the training phase was found to be critical for its performance. The parameterization can be applied online within large-scale atmospheric models, or offline using output from weather forecasting and reanalysis products. Significance Statement Vertical air motion plays a crucial role in several atmospheric processes, such as cloud droplet and ice crystal formation. However, it often occurs at scales smaller than those resolved by standard atmospheric models, leading to uncertainties in climate predictions. To address this, we present a novel deep learning approach that synthesizes data from various sources, providing a representation of small-scale vertical wind velocity suitable for integration into atmospheric models. Our method demonstrates high accuracy when compared to observation-based retrievals, offering potential to mitigate uncertainties and enhance climate forecasting.

PV generation forecasting utilizing a classification-only approach

The increasing use of photovoltaic (PV) systems in electricity infrastructure poses new reliability challenges, as the supply of solar energy is primarily dependent on weather conditions. Consequently, to mitigate the issue, enhanced day-ahead PV production forecasts can be obtained by employing advanced machine learning techniques and reducing the uncertainty of solar irradiance predictions through statistical processing. The objective of this study was to present a methodology for accurately forecasting day-ahead PV production using novel machine learning techniques and a classification-only forecasting approach. Specifically, the central component of the proposed method is a classifier model based on an Extreme Gradient Boosting (XGBoost) ensemble algorithm that classifies the respective daily 30-min profiles of the forecasted global horizontal irradiance (GHI), the measured incident irradiance (Gi), and the AC power (PAC) into a predetermined number of classes. The formed classifier model was used as a dictionary to designate the newly arrived forecasted GHI to a particular class and ultimately identify the corresponding forecasted PAC. The results demonstrated that the proposed forecasting solution provided forecasts with a daily normalised root mean square error (nRMSE) of 8.20% and a mean absolute percentage error (MAPE) of 6.91% over the test set period of one year, while the model's reproducibility was also evaluated and confirmed. Additionally, a comprehensive evaluation based on clear-sky index categories revealed that the model's performance was notably accurate on clear-sky days, while maintaining acceptable accuracy levels on moderate and overcast days. These findings underscore the versatility and robustness of the proposed methodology in handling diverse weather conditions and hold promise for improved PV production forecasts.

Three-dimensional pore-scale study of methane hydrate dissociation mechanisms based on micro-CT images

<p>Methane hydrate is a promising source of alternative energy. An in-depth understanding of the hydrate dissociation mechanism is crucial for the efficient extraction. In the present work, a comprehensive set of pore-scale numerical studies of hydrate dissociation mechanisms is presented. Pore-scale lattice Boltzmann (LB) models are proposed to simulate the multiphysics process during methane hydrate dissociation. The numerical simulations employ the actual hydrate sediment pore structure obtained by the micro-CT imaging. Experimental results of xenon hydrate dissociation are compared with the numerical simulations, indicating that the observed hydrate pore habits evolution is accurately captured by the proposed LB models. Furthermore, simulations of methane hydrate dissociation under different sediment water saturations, fluid flow rates and thermal conditions are conducted. Heat and mass transfer limitations both have significant effects on the methane hydrate dissociation rate. The bubble movement can further influence the dissociation process. Dissociation patterns can be divided into three categories, uniform, non-uniform and wormholing. The fluid flow impacts hydrate dissociation rates differently in three-dimensional real structures compared to two-dimensional idealized ones, influenced by variations in hydrate pore habits and flow properties. Finally, upscaling investigations are conducted to provide the permeability and kinetic models for the representative elementary volume (REV)-scale production forecast. Due to the difference in the hydrate pore habits and dissociation mechanisms, the three-dimensional upscaling results contrast with prior findings from two-dimensional studies. The present work provides a paradigm for pore-scale numerical simulation studies on the hydrate dissociation, which can offer theoretical guidance on efficient hydrate extraction.</p>

Challenges and effectiveness of IT tools used in mining for forecasting production parameters

The research examines the challenges and effectiveness of IT tools used in the mining industry for forecasting production parameters, a crucial aspect as mining operations become increasingly complex. Accurate forecasting is essential not only for optimizing efficiency, safety, and profitability but also for managing the high risks associated with the construction and operation of mining plants, which require long-term financial security and flexible production management. The study explores various IT solutions, such as advanced data analytics, machine learning algorithms, and simulation models, employed to predict key production parameters like ore quality, equipment performance, and resource availability. However, significant challenges, such as data quality issues, the integration of diverse data sources, and the need for specialized expertise, pose obstacles to the effective use of these tools. Despite these challenges, the research finds that IT tools can lead to more accurate forecasting, improved decision-making, and enhanced operational planning, provided that technical and organizational hurdles are addressed. The paper also highlights the importance of modern IT tools in mining production scheduling, demonstrating their advantages over older tools like MS Excel. By presenting examples from coal and copper ore deposits, the study shows that these modern tools not only increase the accuracy of production forecasts but also enable the creation of multiple scenarios and the rapid modification of schedules, which are key to maintaining the competitiveness and liquidity of mining companies.

Prediction of energy production in a building-integrated photovoltaic system using machine learning algorithms

Globally, the building sector is a significant consumer of energy. Implementing effective energy management system (EMS) strategies is crucial for reducing energy consumption and optimizing energy usage across all building elements. Moreover, solar energy is increasingly recognized as essential for sustainable building and urban development. Thus, the energy production forecast from photovoltaic systems installed on buildings is pivotal for a smart EMS's realization and optimal functioning. By incorporating accurate energy production forecasts, we can optimize energy consumption by fitting consumption patterns with production peaks and valleys. This study presents a comparative analysis of five widely used machine learning regressing models, including Random Forest, Decision Trees, XGBoost, LightGBM, and CatBoost for predicting the hourly energy production from a PV system mounted on a building. In this investigation, weather data and time-related characteristics were utilized as predictors. The results showed that LightGBM outperformed the other models, proving its suitability for PV energy forecasting tasks.

Оценка опережающего индикатора ВРП методом темпорального дезагрегирования

Introduction. Тhe estimation of gross regional product high-frequency leading indicator is relevant for the reliable analysis of current trends in regional economy and early understanding of its changes in the periods of high uncertainty since gross regional product is published on an annual basis. One approach to receive this indicator is a temporal disaggregation method, which has proven to be reasonable in foreign literature for disaggregating gross domestic product. At the same time, temporal disaggregation of regional economic time series has been understudied. Purpose. The purpose of the study is to refer to a temporal disaggregation method to estimate an unobservable indicator with high accuracy approximation for annual GRPs. Materials and Methods. The study analyzed Rosstat data for various periods, which characterize the economic growth in the Republic of Bashkortostan, as well as Bank of Russia data of enteprises’ monitoring. X-13ARIMA-SEATS methods for seasonal adjustment, temporal disaggregation methods (Chou–Lin, Fernandez and Litterman) and ARIMA for short-term forecast were used. Results. The article presents the results of temporal disaggregation of the gross regional product of the Republic of Bashkortostan. The best specification was estimated by the Chow–Lin method and includes indicators that characterize industrial production, retail trade, as well as enterprises’ survey data about the fluctuations in the ruble exchange rate. The ARIMA model gave a short-term forecast for a gross regional product leading monthly indicator. Unlike a random walk model with a forecast of up to 2-year lead time, a combination of temporal disaggregation method and ARIMA gave a better out-of-sample annual GRP forecast. Conclusion. The study successfully tested a temporal disaggregation method for the gross regional product of the Republic of Bashkortostan. In practice, this method provides reliable forecast estimates of the gross regional product for the current economic analysis with regard to available high-frequency data. It is shown that the use of survey data can improve the quality of gross regional product forecast.

Climate Yield Models for Some Arable Crops in Ondo State.

This study looks at the connection between easily observed climate variables and crop production. Selected arable crops’ yield statistics spanning ten years were analyzed, including melon, rice, cassava, yam, and maize. Using multiple linear regression models calibrated for each of the selected crops, the crop yield was the dependent variable in SPSS statistical package version 22.0. The meteorological data, including mean temperature (oC), rainfall (mm), evaporation (mm/yr), and relative humidity (%), were correlated with the crop yield. Standard error of estimates (SE), coefficient of determination (R2), and correlation coefficient (R) are the goodness-of-fit parameters that are used to validate the models. The results demonstrated that the R-value for maize is 0.83 with a SE of 4.4 tons/ha, and the R-value for cassava is 0.85 with a SE of 10.2 tons/ha. R-values for yam and melon are 0.77 and 44.5 tons/ha, respectively; for rice and cocoyam, the values were 0.64 and 16.6 tons/ha, respectively. Melon has an R-value of 0.69 and 18.6 tons/ha. Factors such as crop management methods, diseases, nutrients, and other factors that are challenging to account for in the current statistical modeling approach could be the cause of the difference in the goodness-of-fit. There is a great need for efforts to produce trustworthy data for agricultural productivity forecasting and planning in the study region as well as all over the nation.

ARIMA and ARIMAX Analysis on the Effect of Variability of Rainfall, Temperature on Wheat Yield in Haryana

The national and state governments require crop production forecasts to make a variety of policy decisions on import-export, storage, distribution, price, and other factors. This article presents a pre-harvest forecasting method specially developed for crops grown in the western region of Haryana (India). The western region includes Hisar, Sirsa, and Bhiwani districts. For crop forecasting in the Hisar, Sirsa, and Bhiwani regions ARIMA and ARIMAX models have been framed. For the development of the ARIMAX model, climate data during the growing season of the crop were used as input along with the crop yield. The percentage difference between the root mean square error and the wheat yield estimations determined by the real-time yield(s) indicates how well-competing models performed in terms of forecasting. The ARIMAX model performs well at all time points with a lower measurement error compared to the ARIMA model.

Increasing the prospective capacity of global crop and rangeland monitoring with phenology tailored seasonal precipitation forecasts

Droughts are more and more often a limiting factor to agricultural production and can have severe negative effects on food security in vulnerable countries. Global agriculture early warning systems monitor agriculture in near real-time by analyzing meteorological data (e.g. precipitation and temperature) and optical remote sensing data as proxy vegetation health to detect possible negative anomalies and trigger warnings. Seasonal climate forecast can add a predictive component and inform about upcoming precipitation deficits, thus allowing anticipation and improved planning of response actions. Here, we propose a scheme to adapt the standard precipitation forecast from the seasonal Copernicus Climate Change Service multi-system to crop and rangeland phenology, making them suitable for agricultural early warning. Precipitation forecasts are first elaborated into tercile maps showing the probability of the most likely tercile (i.e. drier than normal, normal, wetter than normal) and associated skills of all possible monthly periods combinations included in the six months forecasting horizon. Afterwards, agronomically relevant tercile maps are produced for the closest season in time at any location. These maps are obtained by mosaicking the forecasts for the appropriate growing season period at each grid cell. The resulting map shows the tercile probability for the remaining part of the ongoing growing season (if any at time of analysis) or the probability of the next upcoming season (if in between growing season at time of analysis). The proposed methodology offers a precipitation seasonal forecast product ready to use by agricultural analysts and directly ingestible by automatic warning systems.

Automatic Generation of Artificial Space Weather Forecast Product Based on Sequence-to-sequence Model

Automatic Generation of Artificial Space Weather Forecast Product Based on Sequence-to-sequence Model

From California’s Extreme Drought to Major Flooding: Evaluating and Synthesizing Experimental Seasonal and Subseasonal Forecasts of Landfalling Atmospheric Rivers and Extreme Precipitation during Winter 2022/23

Abstract California experienced a historic run of nine consecutive landfalling atmospheric rivers (ARs) in three weeks’ time during winter 2022/23. Following three years of drought from 2020 to 2022, intense landfalling ARs across California in December 2022–January 2023 were responsible for bringing reservoirs back to historical averages and producing damaging floods and debris flows. In recent years, the Center for Western Weather and Water Extremes and collaborating institutions have developed and routinely provided to end users peer-reviewed experimental seasonal (1–6 month lead time) and subseasonal (2–6 week lead time) prediction tools for western U.S. ARs, circulation regimes, and precipitation. Here, we evaluate the performance of experimental seasonal precipitation forecasts for winter 2022/23, along with experimental subseasonal AR activity and circulation forecasts during the December 2022 regime shift from dry conditions to persistent troughing and record AR-driven wetness over the western United States. Experimental seasonal precipitation forecasts were too dry across Southern California (likely due to their overreliance on La Niña), and the observed above-normal precipitation across Northern and Central California was underpredicted. However, experimental subseasonal forecasts skillfully captured the regime shift from dry to wet conditions in late December 2022 at 2–3 week lead time. During this time, an active MJO shift from phases 4 and 5 to 6 and 7 occurred, which historically tilts the odds toward increased AR activity over California. New experimental seasonal and subseasonal synthesis forecast products, designed to aggregate information across institutions and methods, are introduced in the context of this historic winter to provide situational awareness guidance to western U.S. water managers.

Leveraging Exogenous Insights: A Comparative Forecast of Paddy Production in Nepal Using ARIMA and ARIMAX Models

Using annual time series data from 1975 to 2022, this study analyzed the ARIMA (1,1,7) and ARIMAX (1,1,7) models to improve paddy production forecasts in Nepal. The ARIMA model was initially employed to forecast paddy production. The availability of agricultural land was subsequently included as an exogenous variable in the ARIMAX model (after a significant endogeneity test) to increase precision. In contrast to the ARIMA model, which predicted paddy production of 5787.64 metric tons per hectare for the year 2022, the ARIMAX model predicted 5681.17 metric tons per hectare. Compared to the ARIMA model's MAE (0.0295), MAPE (0.797), and RMSE (0.0373), the ARIMAX model exhibited better accuracy with lower values of MAE (0.0247), MAPE (0.667), and RMSE (0.0301). The results emphasize how crucial it is to include pertinent exogenous variables to capture important dynamics influencing agricultural output. The study has important ramifications for Nepalese policymakers and agricultural planners since it suggests that the ARIMAX model can provide additional insight for policies that maximize production and optimize land usage, hence promoting food security and economic stability.

Forecasting Wind and Solar Energy Production in the Greek Power System using ANN Models

Renewable energy sources (RES) like solar and wind are quite uncertain because of the unpredictable nature of wind and sunlight. As a result, there are at present several issues with system security and the transformed structure of the energy market due to the increasing utilization of renewable energy sources (wind and solar). Accurate forecasting of renewable energy production is extremely important to ensure that the produced energy is equal to the consumed energy. Any deviations have an impact on the system's stability and could potentially cause a blackout in some situations. The issue of the high penetration of RES is discussed in this study along with a novel method of predicting them using artificial neural networks (ANN). The SARIMA prediction model is contrasted with the ANN approach. The suggested ANN for wind power plants has a mean average prediction error (MAPE) of 3%–4.3%, whereas the SARIMA model has a MAPE of 5%–6.5%. In comparison, the present prediction approaches typically have a MAPE of 5%–10%. When the MAPE of solar power plants was calculated, it was also discovered that the SARIMA model had a MAPE of 2.3%–4% and the suggested ANN had a MAPE of 1.4%–2.3%, whereas the MAPE of the present prediction methods was often about 9%.

Application of Internet of Things Technologies in Agriculture

The development of agriculture in the Russian Federation and the Republic of Belarus includes implementing «smart systems» in agriculture based on modern wireless, intelligent technologies and the Internet of Things. This survey presents related works published in the last decade on the use of the Internet of Things to develop agriculture. The survey is based on publications from the scientific electronic library eLIBRARY.ru. We categorized the publications according to the areas of agricultural production as follows: animal husbandry, crop production, greenhouses and weather forecast, water management and irrigation, machinery management, mapping and geodesy, and digital platforms. The survey shows that in Russia and Belarus IoT technologies are developing in agriculture intensively as in advanced countries.

IMPLEMENTASI PERAMALAN PRODUKSI DI PT X DENGAN MENGGUNAKAN GREY METHOD

The stages of a good production process begin with carrying out production planning and forecasting. This stage is carried out to reduce the impact of uncertainty in the future. There are many methods that can be used in forecasting, but there are often obstacles to the lack of data owned to do forecasting. Gray system theory is one of the theories developed to reduce the problem of lack data owned. This research tries to forecast the needs of PT X using the grey method. Based on the results of forecasting using the Grey Method, the method can produce forecasting with a good level of accuracy and is on the criteria of "Qualified" or good enough with an error ratio value of C of 0.36