Forecasting Process Research Articles

A Novel Two-Stage Framework for Mid-Term Electric Load Forecasting

Electric utilities and planners rely heavily on accurate mid-term load projections to effectively schedule maintenance, coordinate load dispatch, and manage fuel reserves. Current approaches that attempt to forecast electric load for multiple horizons in a single stage, however, often result in inaccurate forecasts due to the lack of information for predicting the data at the farthest horizon and propagation of large errors. To address this issue, this paper presents a two-step mechanism for forecasting electric load that utilizes the benefits of both Backpropagation Neural Networks (BPNNs) and Radial Basis Function Neural Networks (RBFNNs). The first stage utilizes a pool of BPNNs to make partial predictions. BPNNs are used as they are well known for their capability of fast and accurate predictions. In the second stage, RBFNNs are used to make complete forecasts. RBFNNs are known for their ability to model non-linear relationships in the data which is crucial in this stage of the forecast process. Additionally, this paper also proposes a unique method for generating additional training samples by incorporating stacking noise, which adds more variability to the data, increasing the generalization capability of the models, resulting in better forecasting performance. The performance of the proposed framework has been investigated with simulation studies and has been validated on actual load data of Madhya Pradesh Power Transmission System, India. The performance comparison with the existing methods in terms of both accuracy and reliability shows that the proposed forecasting framework provides more accurate results.

경기종합지수 단기 예측모형 구축 연구

After the declaration of the ‘Endemic,’ it is crucial to understand the cyclical fluctuations and turning points in the overall economic status of the nation. To promptly and accurately grasp the magnitude and speed of these economic changes, a short-term forecasting analysis was conducted using the composite index of business indicators data from January 2003 to April 2023. In this study, a two-step analysis was performed to propose a standardized forecasting process.
 Firstly, for short-term forecasting of the composite index of business indicators, the results of time series analyses considering only endogenous variables were compared with those incorporating exogenous variables. Subsequently, the study explored machine learning methods, which have been actively discussed recently, and compared and analyzed their prediction performance with traditional time series analysis methods. The analysis revealed that the addition of exogenous variables in time series analysis resulted in higher prediction accuracy compared to analyses considering only endogenous variables. Furthermore, machine learning models exhibited superior prediction performance compared to traditional time series models. As a result of the analysis with the Time-GAN model, which has the best predictive performance, it was found that there would be a rise for 1-2 months due to the growth of private consumption caused by the endemic, but it would decrease again in the 3rd month.

Information system for assessing the informativeness of an epidemic process features

The primary objective of this study is to assess the informativeness of various parameters influencing epidemic processes utilizing the Shannon and Kullback–Leibler methods. These methods were selected based on their foundation in the principles of information theory and their extensive application in machine learning, statistics, and other relevant domains. A comparative analysis was performed between the results acquired from both methods, and an information system was designed to facilitate the uploading of data samples and the calculation of factor informativeness impacting the epidemic processes. The findings revealed that certain features, such as “Chronic lung disease,” “Chronic kidney disease,” and “Weakened immunity,” did not carry significant information for further analysis and hindered the forecasting process, as per the data set examined. The developed information system efficiently supports the assessment of feature informativeness, thereby aiding in the comprehensive analysis of epidemic processes and enabling the visualization of the results. This study contributes to the current body of knowledge by providing specific examples of applying the described algorithmic models, comparing various methods and their outcomes, and developing a supportive tool for analyzing epidemic processes.

A Feature-Informed Data-Driven Approach for Predicting Maximum Flood Inundation Extends

As climate change increases the occurrences of extreme weather events, like flooding threaten humans more often. Hydrodynamic models provide spatially distributed water depths as inundation maps, which are essential for flood protection. Such models are not computationally efficient enough to deliver results before or during an event. To ensure real-time prediction, we developed a feature-informed data-driven forecast system (FFS), which interpreted the forecasting process as an image-to-image translation, to predict the maximum water depth for a fluvial flood event. The FFS combines a convolutional neural network (CNN) and feature-informed dense layers to allow the integration of the distance to the river of each cell to be predicted into the FFS. The aim is to ensure training for the whole study area on a standard computer. A hybrid database with pre-simulated scenarios is used to train, validate, and test the FFS. The FFS delivers predictions within seconds making a real-time application possible. The quality of prediction compared with the results of the pre-simulated physically-based model shows an average root mean square error (RMSE) of 0.052 for thirty-five test events, and of 0.074 and 0.141 for two observed events. Thus, the FFS provides an efficient alternative to hydrodynamic models for flood forecasting.

Behavioral biases and the rational decision-making process of financial professionals: significant factors that determine the future of the financial market

PurposeThis paper describes how financial professionals' behavioral biases influence their financial forecast and decision-making process. Most of the earlier studies are focused on well-developed financial markets, and little is researched about financial professionals, such as institutional investors, portfolio managers, investment advisors, financial analysts, etc., in emerging markets.Design/methodology/approachAn expert-validated questionnaire measure four prominent behavioral biases and Indian financial professionals' rational decision-making process. The final sample consists of 274 valid responses using the purposive sampling technique. IBM SPSS and AMOS structural equation modeling (SEM) software are used to build measurement and structural models, multivariate analysis including regression, factor analysis, etc.FindingsThe results provide empirical insights into the relationship between behavioral biases and the decision-making process. The results suggest that the structural path model closely fits the sample data. The presence of behavioral biases indicates that financial professionals' forecasting and decision-making is not always rational but bounded rational or irrational due to these factors. Furthermore, these biases (except overconfidence bias) have a markedly significant and positive relationship with irrational decision-making.Research limitations/implicationsIt is critical to eradicate these psychological errors, but awareness and attentiveness toward behavioral biases may help financial professionals to make informed decisions. Investors can improve their portfolio decisions and investments by recognizing their judgment errors and focusing on specific investment strategies to mitigate the impact of these biases. It is necessary to incorporate behavioral insights while developing training techniques for financial professionals. Rules of thumb, visual tools, financial coaching and implementing social-cultural elements in training programs enable financial professionals to develop simple, engaging, appealing and customized approaches for their clients.Originality/valueThis novel study is the first of this kind of research that examines the relationship between financial professionals' behavioral biases and rational decision-making process. This study significantly and remarkably provides insights into irrationality in financial professionals' decision-making.

Gaussian Adapted Markov Model with Overhauled Fluctuation Analysis-Based Big Data Streaming Model in Cloud.

An accurate resource usage prediction in the big data streaming applications still remains as one of the complex processes. In the existing works, various resource scaling techniques are developed for forecasting the resource usage in the big data streaming systems. However, the baseline streaming mechanisms limit with the issues of inefficient resource scaling, inaccurate forecasting, high latency, and running time. Therefore, the proposed work motivates to develop a new framework, named as Gaussian adapted Markov model (GAMM)-overhauled fluctuation analysis (OFA), for an efficient big data streaming in the cloud systems. The purpose of this work is to efficiently manage the time-bounded big data streaming applications with reduced error rate. In this study, the gating strategy is also used to extract the set of features for obtaining nonlinear distribution of data and fat convergence solution, used to perform the fluctuation analysis. Moreover, the layered architecture is developed for simplifying the process of resource forecasting in the streaming applications. During experimentation, the results of the proposed stream model GAMM-OFA are validated and compared by using different measures.

Feasibility study of deep learning based on data-based e-commerce operations

Abstract In this paper, we first study data-based e-commerce operation from the big data perspective and elaborate on data-based e-commerce operation from four aspects: professional terminology, operation core, theoretical operation basis, and operation method. Secondly, in the process of e-commerce data-based operation sales prediction, data pre-processing and feature selection are required, and the quality of data and features directly determines the accuracy of the model and based on deep learning, the structure of convolutional and XG fusion prediction model is proposed. Then, three-dimensional data frames are constructed based on four-dimensional data facets, which leads to more accurate model predictiveness, and the prediction model and e-commerce financial and operational risks are studied and analyzed. The results show that the convolutional and XG fusion forecasting model has a greater efficiency enhancement function than the traditional data computation model in the practical application of the forecasting model, proving the model’s stability in the long-term forecasting process. By predicting risk in e-commerce operations in time and improving it in real-time, this study can help the enterprise develop more competitively.

A new stable and interpretable flood forecasting model combining multi-head attention mechanism and multiple linear regression

Abstract This article proposes a multi-head attention flood forecasting model (MHAFFM) that combines a multi-head attention mechanism (MHAM) with multiple linear regression for flood forecasting. Compared to models based on Long Short-Term Memory (LSTM) neural networks, MHAFFM enables precise and stable multi-hour flood forecasting. First, the model utilizes characteristics of full-batch stable input data in multiple linear regression to solve the problem of oscillation in the prediction results of existing models. Second, full-batch information is connected to MHAM to improve the model's ability to process and interpret high-dimensional information. Finally, the model accurately and stably predicts future flood processes through linear layers. The model is applied to Dawen River Basin, and experimental results show that the MHAFFM, compared to three benchmarking models, namely, LSTM, BOA-LSTM (LSTM with Bayesian Optimization Algorithm for Hyperparameter Tuning), and MHAM-LSTM (LSTM model with MHAM in hidden layer), significantly improves the prediction performance under different lead time scenarios while maintaining good stability and interpretability. Taking Nash–Sutcliffe efficiency index as an example, under a lead time of 3 h, the MHAFFM model exhibits improvements of 8.85, 3.71, and 10.29% compared to the three benchmarking models, respectively. This research provides a new approach for flood forecasting.

Profit Forecast Accuracy of Time Series Model - A Case Study of Associated British Foods

Accurate earnings per share (EPS) forecasting is crucial for financial decision-making. This study explores the potential of improving EPS forecasting accuracy by integrating economic lead indicators into time-series models. By incorporating macroeconomic factors like GDP growth and interest rates, the models capture the influence of the broader economic environment on a company’s financial performance. Results demonstrate that including economic lead indicators significantly enhances EPS predictability beyond traditional time-series models. This integration offers a forward-looking perspective, comprehensive analysis, and context to the forecasting process, enabling stakeholders to make more informed investment decisions and develop better strategies. Further research can investigate additional lead indicators, assess their impact in different industries, and develop hybrid forecasting models for refined EPS predictions.

Transformer-based forecasting for intraday trading in the Shanghai crude oil market: Analyzing open-high-low-close prices

The Shanghai crude oil futures market exudes distinct speculative attributes, underscoring the pivotal significance of precise price forecasts. Accurate forecasting of Shanghai crude oil futures prices assumes vital importance for investors to optimize their portfolios profitably, for producers to mitigate production risks in the crude oil spot market, and for providing cogent decision-making support to government entities. This study implements a groundbreaking unbiased structural forecasting for Shanghai crude oil futures' open-high-low-close (OHLC) prices leveraging the Transformer framework coupled with the model-driven and penalty term-based loss function designs. Based on OHLC forecasts, this study devises three intraday trading strategies. Notably, our results evince that the forecasting accuracy of the Transformer outperforms the Naïve method, vector autoregression (VAR) and vector error correction model (VECM), multiple linear regression (MLR), support vector regression (SVR), and long short-term memory (LSTM) neural network when applied to different temporal granularities of Shanghai crude oil futures OHLC data. Furthermore, the three proposed intraday trading strategies exhibit higher annualized return rates and Sharpe ratios, alongside lower maximum drawdowns and standard deviations of returns in comparison to the conventional close-to-close strategy relying solely on the close price. Remarkably, the forecasting process and intraday trading strategies explicated in this study can equally apply to other futures products in the energy sector, including electricity, coal, and natural gas.

Harnessing eXplainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of Grad-CAM and SHAP

This study investigates the efficacy of Explainable Artificial Intelligence (XAI) methods, specifically Gradient-weighted Class Activation Mapping (Grad-CAM) and Shapley Additive Explanations (SHAP), in the feature selection process for national demand forecasting. Utilising a multi-headed Convolutional Neural Network (CNN), both XAI methods exhibit capabilities in enhancing forecasting accuracy and model efficiency by identifying and eliminating irrelevant features. Comparative analysis revealed Grad-CAM’s exceptional computational efficiency in high-dimensional applications and SHAP’s superior ability in revealing features that degrade forecast accuracy. However, limitations are found in both methods, with Grad-CAM including features that decrease model stability, and SHAP inaccurately ranking significant features. Future research should focus on refining these XAI methods to overcome these limitations and further probe into other XAI methods’ applicability within the time-series forecasting domain. This study underscores the potential of XAI in improving load forecasting, which can contribute significantly to the development of more interpretative, accurate and efficient forecasting models.

Digital Maturity of Forecasting and its Impact in Times of Crisis

AbstractEspecially in times of crisis, reliable predictions about probable future developments are difficult, but critical for successfully managing business operations. At the same time, it remains unclear what constitutes a good forecasting process during crises. The aim of this study is to analyze whether and how digital transformation can enhance forecasting processes and enable firms to better deal with crises. To do so, we refer to the concept of digital maturity, i.e., the extent to which digital transformation is adopted in internal processes, studied at the practice of forecasting. Specifically, we analyze whether digitally more mature forecasting processes positively influence (1) satisfaction with forecasting during crises, (2) the effectiveness of countermeasures, and (3) the economic situation during crises. We conduct a cross-sectional survey among 195 medium-sized and large companies in Germany to shed light on the forecasting process and its digital maturity as well as on the impact of the COVID-19 economic crisis on companies. Based on ordinary least squares (OLS) regression, we find that digitally more mature forecasts increase satisfaction with forecasting and the effectiveness of countermeasures. Overall, this study provides new insights into relevant aspects of forecasting to support successful crisis management, and it highlights the importance of advancing digital transformation in forecasting, especially to successfully deal with crises.

A Study on Role of Financial Planning and Forecasting in Organizational Growth and Development

The group is made up of a number of individuals who must cooperate in order to accomplish the stated goals. The business must create and present a strategy that outlines the direction and establishes goals for this reason. An estimate's purpose is to reveal what resources might be required for the business to effectively implement its strategy. Organizations can manage liabilities, investment possibilities, and cash flow with the aid of financial planning. Setting objectives, forecasts, resource requirements, and other indicators in order to achieve them within a reasonable budget is the process of financial planning and forecasting. The projections show how much money the company will need from investors or consumers to meet its strategic goals. One component of management accounting, which encompasses all facets of financial reporting having data on assets, liabilities, and financial results, is corporate finance. The importance of planning and forecasting for Indian multinational corporations is the topic of the present study. The research was conducted based on a survey of 200 employees from the finance departments of ten major MNCs with offices in Pune and Mumbai. The study's findings suggest that Indian multinational corporations (MNCs) are eager to improve their financial projections and planning so they can make better choices that will help them reach their objectives.

A hybrid forecasting model with logistic regression and neural networks for improving key performance indicators in supply chains

This study investigates the potential of predictive analytics in improving Key Performance Indicators (KPIs) forecasting by leveraging Lean implementation data in supply chain enterprises. A novel methodology is proposed, incorporating two key enhancements: using Lean maturity assessments as a new data source and developing a hybrid forecasting model combining Logistic regression and Neural Network techniques. The proposed methodology is evaluated through a comprehensive empirical study involving 30 teams in a large supply chain company, revealing notable improvements in forecasting accuracy. Compared to a baseline scenario without process improvement data, the new methodology achieves an enhanced accuracy score by 17% and an improved F1 score by 13 %. These findings highlight the benefits of integrating Lean maturity assessments and adopting a hybrid forecasting model, contributing to the advancement of supply chain analytics. By incorporating lean maturity assessments, the forecasting process is enhanced, providing a deeper comprehension of the underlying Lean framework and the impact of its elements on supply chain performance. Additionally, adopting a hybrid model aligns with current best practices in forecasting, allowing for the utilisation of various techniques to optimise KPI prediction accuracy while leveraging their respective strengths.

Short-Term (7 Day) Beaufort Sea Ice Extent Forecasting with Deep Learning

Abstract Ships inside the Arctic basin require high-resolution (1–5 km), near-term (days to semimonthly) forecasts for guidance on scales of interest to their operations where forecast model predictions are insufficient due to their coarse spatial and temporal resolutions. Deep learning techniques offer the capability of rapid assimilation and analysis of multiple sources of information for improved forecasting. Data from the National Oceanographic and Atmospheric Administration’s Global Forecast System, Multi-scale Ultra-high Resolution Sea Surface Temperature (MEaSUREs), and the National Snow and Ice Data Center’s Multisensor Analyzed Sea ice Extent (MASIE) were used to develop the sea ice extent deep learning forecast model, over the freeze-up periods of 2016, 2018, 2019, and 2020 in the Beaufort Sea. Sea ice extent forecasts were produced for 1–7 days in the future. The approach was novel for sea ice extent forecasting in using forecast data as model input to aid in the prediction of sea ice extent. Model accuracy was assessed against a persistence model. While the average accuracy of the persistence model dropped from 97% to 90% for forecast days 1–7, the deep learning model accuracy dropped only to 93%. A k-fold (fourfold) cross-validation study found that on all except the first day, the deep learning model, which includes a U-Net architecture with an 18-layer residual neural network (Resnet-18) backbone, does better than the persistence model. Skill scores improve the farther out in time to 0.27. The model demonstrated success in predicting changes in ice extent of significance for navigation in the Amundsen Gulf. Extensions to other Arctic seas, seasons, and sea ice parameters are under development. Significance Statement Ships traversing the Arctic require timely, accurate sea ice location information to successfully complete their transits. After testing several potential candidates, we have developed a short-term (7 day) forecast process using existing observations of ice extent and sea surface temperature, with operational forecasts of weather and oceanographic variables, and appropriate machine learning models. The process included using forecasts of atmospheric and oceanographic conditions, as a human forecaster/analyst would. The models were trained for the Beaufort Sea north of Alaska using data and forecasts from 2016 combined with 2018–20. The results showed improvement in short-term forecasts of ice locations over current methods and also demonstrated correctly predicted changes in the sea ice that are important for navigation.

The NOAA Weather Prediction Center’s Use and Evaluation of Experimental Warn-on-Forecast System Guidance

This study examines use of experimental Warn-on-Forecast System (WoFS) guidance for short-term flash flood prediction at the NOAA Weather Prediction Center’s Meteorological Watch (Metwatch) desk. The WoFS guidance provides storm-scale ensemble forecasts for individual thunderstorms out to six hours and has previously shown great promise in its predictive skill for heavy rainfall events. Its operational utility was examined during 2019 and 2020 in a formal collaboration between Warn-on-Forecast scientists and Metwatch meteorologists. During that time, Metwatch meteorologists integrated real-time WoFS guidance into their Mesoscale Precipitation Discussion forecast processes and provided evaluations via a post-event survey. The survey queried impacts of WoFS guidance on their situational awareness, workload, and confidence, and Metwatch meteorologists also reported subjective assessments of model performance. Survey results highlighted the importance of viewing consistency in WoFS guidance across runs and agreement between WoFS guidance with conceptual models, other numerical weather prediction guidance, and observations. The use of WoFS tended to either maintain or slightly increase Metwatch meteorologists’ workload, while also increasing their confidence (notably for events perceived as better predicted). Of the different forecast attributes evaluated, Metwatch meteorologists reported convective mode as the attribute best predicted by WoFS. Use of WoFS guidance supported Mesoscale Precipitation Discussion decision making, including the placement and spatial extent of the product and the level of specificity provided about the related flash flood threat(s).

Assessing the impact of big data analytics on decision-making processes, forecasting, and performance of a firm

There are various kinds of applications of BDA in the firms. Not many studies are there which deal with the impact of BDA towards issues like forecasting, decision-making, as well as performance of the firms simultaneously. So, there exists a gap in the research. In such a background, this study aims at examining the impacts of BDA on the process of decision-making, forecasting, as well as firm performance. Using resource-based view (RBV) as well as dynamic capability view (DCV) and related research studies, a research model was proposed conceptually. This conceptual model was validated taking help of PLS-SEM approach considering 366 respondents from Indian firms. This study has highlighted that smart decision making and accurate forecasting process can be achieved by using BDA. This research has demonstrated that there is a considerable influence of adoption of BDA on decision making process, forecasting process, as well as overall firm performance. However, the present study suffers from the fact that the study results depend on the cross-sectional data which could invite defects of causality and endogeneity bias. The present research work also found that there is no impact of different control variables on the firm's performance.

Improving Air Quality Prediction with a Hybrid Bi-LSTM and GAN Model

Air quality is a topic that has been of utmost concern across the globe for the past few decades. Various intelligent monitoring systems are used in diverse scenarios, collecting air quality data that contains missing values. Such missing values in data cause hindrances in forecasting. This time series prediction or forecasting process extracts the necessary information from historical records and predicts future values. To solve the missing values issue in data, Generative Adversarial Networks (GAN) are used to impute the missed data. While the learning of long-term dependencies embedded in the time series poses another threat to the models in the time prediction. To overcome this, Long Short-Term Memory (LSTM) models are used. Yet, most of the neural network-based methods failed to consider the patterns of time series data that varied for each period, and the encoder-decoder performance deteriorated for longer sequences. To combat this, the present study proposes a hybrid probabilistic model to generate parameters for predictive distribution at every step. Hence, an implementation of hierarchical-attention-based BiLSTM with GAN is proposed in the study for effective prediction and minimal error. The proposed model is assessed with the evaluation metrics such as Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and Mean Square Error (MSE). The evaluation metric confirmed the higher accuracy of the proposed model than the existing models in time series prediction.

An IoT based forest fire detection system using integration of cat swarm with LSTM model

The destruction of millions of acres of forest each year by forest fires is a global environmental crisis that has real-world consequences for people's livelihoods and the health of our planet. The ability to foresee the onset of such a natural disaster is, thus, of paramount importance in reducing this risk. There have been numerous proposed technologies and novel approaches for detecting and preventing forest fires. Integrating AI to automate fire prediction and detection is becoming increasingly common. To provide effective forest fire detection, people make use of several technological expansions, with the IoT for data collecting and Artificial Intelligence (AI) for the forecast process. Artificial intelligence (AI) is a key study technique that has been proven to be the best in enhancing the presentation of detecting fire threats in important locations by several researchers. Due to the importance of object detection in this investigation, EfficientDet was chosen for implementation. It is suggested that fire breakouts be detected using a Recurrent LSTM Neural Network (RLSTM-NN). Here, we propose a Cat Swarm Fractional Calculus Optimization (CSFCO) algorithm for deep learning that combines the best features of Cat Swarm Optimization (CSO) with fractional calculus for optimal training results (FC). Terms of the simulation results reveal that the suggested process outdoes the state-of-the-art approaches. The suggested typical can identify the onset of a fire with a precision of 98.6% and an error rate of only 0.14%.

GEO-INFORMATION ANALYSIS OF MUDFLOW DANGER AND THREAT ASSESSMENT FOR BRIDGE STRUCTURES WITHIN THE TERRITORY OF TRANSCARPATHIA

Problem Statement and Purpose. Mudflows are quite common in the world and in Ukraine in particular. There are three mudflow-prone basins in the Carpathian region of Ukraine. Mudflowing processes threaten infrastructure facilities, in particular highways and bridge crossings. Therefore, it is advisable to monitor and forecast their development. Geoinformation technologies are the main tool that should be used for monitoring and modeling the development of mudflow processes. Damage to bridges, which are primarily exposed to the mudflow process, as a result of mudflows, is caused by phenomena that can be summarized by the terms – impact, abrasion, erosion and vibration. Mudflows also activate erosion phenomena, change the position of the watercourse channel, which leads to damage to bridge structures. The purpose of the study is to assess the risk of mudflows for bridge structures for the territory of eastern Transcarpathia using methods of geoinformation analysis and modeling. Data & Methods. At the present stage, it is common to model and forecast the development of mudflow processes based on taking into account the combined effect of factors. The main factors of the occurrence and development of mudflow phenomena include: tectonic, geological, hydrological conditions, modern exogenous geological processes, soil and plant cover and anthropogenic activity. The process of modeling and forecasting of mudflow hazard includes such basic stages as the selection of a set of factors affecting the development of mudflow, assessment of the level of their influence, calculation of integral indices of sediment activity, creation of a prognostic geo-informational model of mudflow hazard. Results. A spatial prognostic model of mudflow hazard was created for the area that occupies the eastern part of Transcarpathia, the main part of which is located in the mountainous Carpathians. Mudflow processes of rain genesis are mainly observed in the mentioned territories. The spatial model has a constant character, since the spatial factors of the development of mudflow processes are not variable in time and determine the probability of the occurrence of a mudflow phenomenon in space, the period of activation is determined by the influence of time factors for which a time complex indicator should be calculated. Based on the created mudflow hazard model for the research area, the mudflow hazard for bridge crossings was assessed. It was established that a high probability of mudflow hazard is characteristic of 45 bridges. These bridge structures are located in the upper reaches of the Tereblya and Teresva rivers, in the basin of the Black Tysa and White Tysa rivers. For a detailed assessment of engineering and geological threats to bridge structures as a result of the development of mudflows, floods, coastal erosion processes, it is advisable to use a comprehensive approach, using the toolkit of geoinformation systems and a wide range of data sources, in particular, data from remote sensing of the Earth.