Modeling Process Research Articles

Leveraging social media data in knowledge management to identify noncompliance: insights from the foodservice industry

Purpose Social media data contains a wealth of content related to customers’ reactions to, and comments on, firms’ performance. Through the lens of signaling theory, this paper aims to investigate the use of social media data as a knowledge resource in communicating firms’ noncompliance risk to regulatory agencies. Design/methodology/approach This paper proposes a two-step social media analytics framework to detect noncompliant firms. First, it creates a context-specific dictionary that contains keywords relevant to firms’ noncompliant behaviors. Next, it extracts those keywords from customer reviews, customer sentiment and emotions to predict firm noncompliance. It tests these ideas in the context of food safety regulations. Findings It identified over 100 words that are related to restaurants’ hygiene deficiencies. Using the occurrence of these words in customer reviews, as well as sentiments and emotions expressed within them, the author’s best-performing model can identify nearly 90% of the restaurants that severely violated regulations. Practical implications After being processed by appropriate machine learning algorithms, customer reviews serve as valuable knowledge resources, enabling regulatory agencies to identify noncompliant firms. Regulatory agencies can use this model to complement the current compliance monitoring scheme. Originality/value This research contributes a novel methodology for creating a context-specific dictionary that keeps only the relevant words customers use when discussing firms’ noncompliant acts. In the absence of such an approach, numerous irrelevant signals would be included in the modeling process, thereby increasing the cost of social media analytics.

Anomaly Detection in Industrial Processes: Supervised vs. Unsupervised Learning and the Role of Explainability

Background Anomaly detection is vital in industrial settings for identifying abnormal behaviors that suggest faults or malfunctions. Artificial intelligence (AI) offers significant potential to assist humans in addressing these challenges. Methods This study compares the performance of supervised and unsupervised machine learning (ML) techniques for anomaly detection. Additionally, model-specific explainability methods were employed to interpret the outputs. A novel explainability approach, MLW-XAttentIon, based on causal reasoning in attention networks, was proposed to visualize the inference process of transformer models. Results Experimental results revealed that unsupervised models perform well without requiring labeled data, offering significant promise. In contrast, supervised models demonstrated greater robustness and reliability. Conclusions Unsupervised ML techniques present a feasible, resource-efficient option for anomaly detection, while supervised methods remain more reliable for critical applications. The MLW-XAttentIon approach enhances interpretability of transformer-based models, contributing to trust and transparency in AI-driven anomaly detection systems.

Cauchy–Logistic Unit Distribution: Properties and Application in Modeling Data Extremes

This manuscript deals with a novel two-parameter stochastic distribution, obtained by transforming the Cauchy distribution, using generalized logistic mapping, into a unit interval. In this way, according to the well-known properties of the Cauchy distribution, a unit random variable with significantly accentuated values at the ends of the unit interval is obtained. Therefore, the proposed stochastic distribution, named the Cauchy–logistic unit distribution, represents a stochastic model that may be suitable for modeling phenomena and processes with emphasized extreme values. Key stochastic properties of the CLU distribution are examined, such as moments, entropy, modality, and symmetry conditions. In addition, a quantile-based parameter estimation procedure, an asymptotic analysis of the thus obtained estimators, and their Monte Carlo simulation study are conducted. Finally, the application of the proposed distribution in stochastic modeling of some real-world data with emphasized extreme values is provided.

Methodology for the Automatic Generation of Optimization Models of Systems of Flexible Energy Resources

The integration of increasing shares of intermittent renewable energy necessitates flexibility in both energy generation and consumption. Typically, the operation of flexible energy resources is orchestrated through optimization models. However, the manual creation of these models is a complex and error-prone task, often requiring the expertise of domain specialists. This work introduces a methodology for the automatic generation of optimization models for systems of flexible energy resources to simplify the modeling process and increase the use of energy flexibility. This methodology utilizes a modular, generic model structure designed to depict systems of flexible energy resources. It incorporates algorithms for model parameter derivation from operational data and an information model that represents the system’s structure and dependencies of resources. The efficacy of this methodology is demonstrated in two case studies, highlighting its relevance and ability to significantly streamline the optimization modeling process by minimizing the need for manual intervention.

Influencing Factors Analysis and Prediction of Gas Emission in Mining Face

Mine gas emission is one of the main causes of gas disasters. In order to achieve the accurate prediction of gas emission, a gas emission prediction model based on the random forest (RF) method was proposed in combination with the analysis of its influencing factors. The prediction results were compared with the support vector regression (SVR) and BP neural network (BPNN) methods, and then they were verified and analyzed through the Dongqu coal mine. The results show that the gas emission prediction model based on random forest has strong generalization and robustness, and RF has a wide range of parameter adaptation during the modeling process. When the number of trees (ntree) exceeds 100, its training error tends to stabilize, and changes in ntree have no substantial impact on the prediction performance. The SVR prediction model has significant bias in both the training and testing stages. Meanwhile, the BPNN model has excellent prediction results in the training phase, but there is a large error in the testing stage, which indicates that there is an “overfitting” phenomenon in the training stage, resulting in weak generalization. The evaluation of variable importance shows that the extraction rate, coal seam depth, daily production, gas content in adjacent layers, and coal seam thickness have a significant impact on gas emission. Meanwhile, through application analysis, it is further demonstrated that the random forest method has high accuracy, strong stability, and universality, and it can achieve good predictive performance without the need for complex parameter settings and optimization, making it is very suitable for predicting gas emission.

A Case Study on Multi-Real-Option-Integrated STO-PF Models for Strengthening Capital Structures in Real Estate Development

This study examines the integration of multi-real-option valuation and security token offering (STO) as an innovative approach to real estate project financing. The case study of Aspen Resort Development serves to illustrate this methodology. The traditional discounted cash flow (DCF) method is frequently ill-suited to the dynamic and uncertain nature of long-term real estate projects, particularly in regard to the ability to adapt to market fluctuations. In order to address these limitations, this study employs a multi-real-option model with a binomial lattice framework, thereby facilitating flexible decision-making in various investment stages. The analysis demonstrates that the STO-based project financing (STO-PF) model offers enhanced financial performance and strategic advantages in comparison to the conventional DCF approach. Furthermore, the STO-PF model has the effect of increasing liquidity, expanding investment accessibility, and improving risk management through the utilization of digital platforms. By quantifying the project’s extended net present value (ENPV), the integration of STOs with real-options models can facilitate optimal investment decisions in the context of a high level of market volatility. Consequently, the STO-PF model is determined to yield a project value (E) of USD 7.34 million and a real-options value (ROV) of USD 3.69 million. This is markedly higher than the net present value (NPV) of USD 3.65 million derived from the traditional project finance (PF) model. Furthermore, the put option for the second investment stage contributes USD 16.45 million to the overall value of the project, thereby demonstrating the flexibility and strategic advantages of the STO framework in comparison to static NPV analysis. The Aspen project serves as a case study, demonstrating the financial viability of phased investments in dynamic market conditions. It contributes to the theoretical understanding of STO-based financing and provides practical insights for developers seeking flexible and innovative financing solutions in the real estate sector. Further research is required to confirm the applicability of STOs in diverse market environments and regulatory contexts. Additionally, in-depth research is necessary to integrate emerging technologies, such as artificial intelligence and machine learning, into multi-real-option-based financial platforms. This integration aims to enhance financial modeling and decision-making processes, as well as to facilitate the integration of digital technologies in this field. Only then can the development and implementation of smart construction development advance.

Spatial-situational approach to airport transport security management

The trend in the development of systems for ensuring and managing aviation security of transport facilities in recent years has acquired the character of a transition from classical regulatory management schemes to automated control schemes using procedures based on the theory of optimal control. A number of factors can be identified that reduce the effectiveness of security management in the current system: problems related to time, post-factum processing of situational events, non-systemic collection and processing of information, the problem of hypothetical and potential security threats, the problem of security integrality and security systems integrality, the problem of indivisibility of security and the problem of vulnerability of transport infrastructure and, finally, the problems of forecasting the development of negative events. All this implies a transition to information security management, which is based on a spatial-situational approach. The paper substantiates the emergence of threat space, protection space and security space and shows the possibilities of using these spaces to solve the problems of formalization of security management procedures in the modeling process.

Construction of intelligent coal blending system based on BP neural network and genetic algorithm

The quality of coke is crucial to the steel-making process, and advanced coal blending technology can effectively improve the stability of coke and reduce production costs. However, some of the current coal blending systems lack expertise and are out of touch with the actual situation of the factory, unable to meet production requirements. In the research process, this paper combined the conventional indicators measured in Chinese factories with actual data, and established a reliable database. In the back propagation neural network modeling process, several parameters were optimised, significantly improving the prediction performance compared to traditional regression methods. An evolutionary algorithm with a penalty matrix is used to optimise the coal blending ratio, resulting in a reduction in the ash and sulphur content of the coke, as well as a cost reduction of approximately 44 RMB per tonne. Based on the initial research results, an intelligent coal blending system was developed, which has been successfully applied in the production process of a coke plant, with simple operation and stable operation, providing important reference for the development of coal blending systems in other coke plants.

Data-Driven Decision Support to Guide Sustainable Grazing Management

Data-driven decision support can help guide sustainable grazing management by providing an accurate estimate of grazing capacity, in coproduction with managers. Here, we described the development of a decision support model to estimate grazing capacity and illustrated its application on two sites in the western United States. For the Montgomery Pass Wild Horse Territory in California and Nevada, the upper limit estimated in the capacity assessment was 398 horses and the current population was 654 horses. For the Eagle Creek watershed of the Apache–Sitgreaves National Forest of eastern Arizona, the lower end of capacity was estimated at 1560 cattle annually, compared to the current average of 1090 cattle annually. In addition to being spatio-temporally comprehensive, the model provides a repeatable, cost-effective, and transparent process for establishing and adjusting capacity estimates and associated grazing plans that are supported by scientific information, in order to support livestock numbers at levels that are sustainable over time, including levels that are below average forage production during drought conditions. This modeling process acts as a decision support tool because it enables different assumptions to be used and explored to accommodate multiple viewpoints during the planning process.

Модель формирования профессиональной надежности курсантов авиационных военных вузов

The purpose of the study is to develop a model for the formation of professional reliability of cadets of aviation military universities. The article considers pedagogical modeling as a means of creating structured ideas about the formation of professional reliability among cadets in the educational process of an aviation military university. In conditions of high responsibility and risk, it is necessary to develop stable qualities in future officers that contribute to their successful professional activities, including in special flight situations, which is an important aspect of training, since flight safety and successful performance of official tasks depend on it. The author suggests a structural and functional model consisting of four components: motivational-predictive, methodological, technological and performance-evaluation, presented in the form of blocks. Each block covers important aspects of the process of forming the professional reliability of cadets. The blocks include planning, selection of strategies and methods, as well as diagnosis and evaluation of results. The scientific novelty of the study lies in the fact that the author has for the first time developed a pedagogical model for the formation of professional reliability of cadets of aviation military universities, which reflects the methodological, structural, technological and evaluative capabilities of the phenomenon under consideration. As a result of the research, the concept of "professional reliability of cadets of aviation military universities" is defined; the potential of the educational environment of a military university is revealed; the essence of the modeling process is determined; the available pedagogical models are analyzed, and a model for the formation of professional reliability of cadets of aviation military universities is presented. The article emphasizes the importance of an integrated approach to the training of flight specialists, taking into account modern requirements for military education and the specifics of the aviation field.

Optimising the Design of a Hybrid Fuel Cell/Battery and Waste Heat Recovery System for Retrofitting Ship Power Generation

This research aims to assess the integration of different fuel cell (FC) options with battery and waste heat recovery systems through a mathematical modelling process to determine the most feasible retrofit solutions for a marine electricity generation plant. This paper distinguishes itself from existing literature by incorporating future cost projection scenarios involving variables such as carbon tax, fuel, and equipment prices. It assesses the environmental impact by including upstream emissions integrated with the Energy Efficiency Existing Ship Index (EEXI) and the Carbon Intensity Indicator (CII) calculations. Real-time data have been collected from a Kamsarmax vessel to build a hybrid marine power distribution plant model for simulating six system designs. A Multi-Criteria Decision Making (MCDM) methodology ranks the scenarios depending on environmental benefits, economic performance, and system space requirements. The findings demonstrate that the hybrid configurations, including solid oxide (SOFC) and proton exchange (PEMFC) FCs, achieve a deduction in equivalent CO2 of the plant up to 91.79% and decrease the EEXI and the average CII by 10.24% and 6.53%, respectively. Although SOFC-included configurations show slightly better economic performance and require less fuel capacity, the overall performance of PEMFC designs are ranked higher in MCDM analysis due to the higher power density.

Selective modal excitation in a multimode nanoslit by interference of surface plasmon waves.

Interference of surface plasmons has been widely utilized in optical metrology for applications such as high-precision sensing. In this paper, we introduce a surface plasmon interferometer with the potential to be arranged in arrays for parallel multiplexing applications. The interferometer features two grating couplers that excite surface plasmon polariton (SPP) waves traveling along a gold-air interface before converging at a gold nanoslit where they interfere. A key innovation lies in the ability to tune the interference pattern by altering the geometrical properties of the gold nanoslit such that one, two or more resonance modes are supported in the nanoslit. Our experimental results validate the approach of our design and modelling process, demonstrating the potential to fine-tune geometrical parameters such as grating coupler pitch, depth, duty cycle, and nanoslit dimensions to alter the transmitted radiation pattern and the transmittance. We demonstrate the ability of a grating coupler to induce focusing of SPP waves to an arbitrary location on chip by illuminating with a converging Gaussian beam. Additionally, we observed far-field interference patterns linked to the multimodal operation of the nanoslit.

A Novel Mechanism-Equivalence-Based Tweedie Exponential Dispersion Process for Adaptive Degradation Modeling and Life Prediction

A Novel Mechanism-Equivalence-Based Tweedie Exponential Dispersion Process for Adaptive Degradation Modeling and Life Prediction

A prognostic index for advanced-stage extranodal natural killer/T-cell lymphoma: A multicenter study.

Advanced-stage extranodal natural killer/T-cell lymphoma (ENKTL) is a highly heterogeneous disease with very poor prognosis. All commonly utilized prognostic models incorporated both early-stage and advanced-stage patients in the modeling process. This study aim to design a prognostic model specifically for advanced-stage ENKTL, providing risk stratification in affected patients. We analyzed 291 patients with stage III/IV ENKTL receiving asparaginase-based chemotherapy from 8 institutions to develop a new prognostic model and validate it in an independent cohort consisted of 221 patients from 4 additional hospitals. The prognostic model included three independent variables based on a multivariate analysis for overall survival (OS): age, bone marrow invasiveness and visceral organ involvement. We identified three different risk groups: group 1, no adverse factors; group 2, one factor; and group 3, two or three factors, which were associated with 5-year OS rates of 66.0%, 32.3%, and 20.0%, respectively (P < 0.001). The prognostic index of natural killer lymphoma (PINK) and nomogram-revised risk index (NRI) were unsatisfactory for stratifying these patients. These results were validated and confirmed in an independent cohort. This newly proposed model can be used to guide risk-adapted treatment for advanced stage ENKTL.

A grinding force prediction model and experimental validation for ultrasonic-assisted end grinding of 2.5D C/SiC composites

In this work, a grinding force predictive model is proposed for processing 2.5D carbon fibre-reinforced silicon carbide matrix composites (C/SiCs) using ultrasonic-assisted end grinding (UAEG). The coupling effects of process parameters on prediction results are explored, and the influence of processing parameters on forces is analyzed from the theoretical perspective. In the modelling process, the anisotropy of the 2.5D C/SiCs, the motion path of the end-face grits, and the grit-workpiece interaction stages were fully considered. According to the material’s removal properties, the interaction of the grit-workpiece was classified into different stages. The proposed model can be derived by considering the forces at different contact stages. To evaluate the model’s reliability, single-factor experiments were conducted to assess its prediction accuracy. The results of validation with experiments demonstrated that the maximum error of the model was less than 12 %, and the average error was only 6.23 %.

Artificial Intelligence-Enhanced Green Building Design for Environmental Sustainability

Green buildings are those that use sustainable methods of construction to either maintain or improve the local quality of life. Decisions affecting a project's quality, safety, profitability, and timetable are made using Artificial Intelligence (AI) in Green Construction by analyzing data gathered from monitoring the construction site and using predictive analytics. For instance, increased accuracy in weather predictions might lead to more production, less waste, lower costs, and less greenhouse gas emissions. Green building construction is a significant source of carbon dioxide released through the breakdown of carbonates. Researchers have concluded that integrating industrial wastes is crucial in green concrete making due to its benefits, such as reducing the requirement for cement. When planning with concrete, its compressive strength must be considered. Due to their high predictive power, AI algorithms may be used to determine the compressive strength of concrete mixtures. Existing artificial intelligence (AI) models may be evaluated for their modeling process and accuracy to inform the creation of new models that more accurately represent the comprehensive evaluation of setting parameters on model performance and boost accuracy. Potential sources of conflict in this anthropocentric future include climate change and the availability of renewable energy sources. Scientists think there is a connection between the increased emission of greenhouse gases like carbon dioxide (Co2) from the combustion of fossil fuels and the acceleration of climate change and global warming. Research has demonstrated that the building sector is a significant source of atmospheric carbon dioxide (Co2). Construction, building activities, and subpar energy sources have all significantly increased atmospheric CO2. The proposed research set out to measure how well AI in Green Building Construction (AI-GBC) might reduce carbon emissions and utility bills. Artificial intelligence uses SVM and GA to reduce energy use and carbon dioxide emissions. Several statistical metrics, such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Root Mean Squared Log Error (RMSLE), are used to evaluate the AI-GBC's precision. Both Machine Learning (ML) models yielded positive results, with prediction accuracies above 95%. Regarding predicting Co_2, GA models were close to the mark, with an R2 of 0.95. Ninety-six percent will complete a performance analysis, and 97% will conduct a k-fold cross-validation analysis. Cross-validation is used to ensure that the findings of the extended modeling technique are accurate and prevent overfitting.

Predicting cobalt ion concentration in hydrometallurgy zinc process using data decomposition and machine learning.

Solid waste is one of the primary contributors to environmental pollution currently, it is crucial to enhance the prevention and control of solid waste pollution in environmental management. The effectiveness of the second stage of purification in the industrial zinc hydrometallurgy is determined by the concentration of cobalt ion. Manual testing and monitoring of cobalt ion concentration are time consuming and costly, and prone to delays, which can result in discharge of cobalt ion concentration that does not meet the standards, leading to water pollution. Additionally, over-addition of zinc powder leads to a waste of resources, increasing the production cost of the company. Here, this work proposes a hybrid prediction model that combines the advantages of data decomposition and machine learning algorithms to predict the metal cobalt ion concentration in the effluent solution of a section of zinc hydrometallurgy refining purification in factory A. According to the different types of experiments, ablation experiments and contrast experiments are designed in this work under the same training and test data were used in the modeling process. Analytic and experimental results show that the proposed hybrid prediction model has the smallest error and the best fit between the actual and predicted values of cobalt ion concentration, and the appropriate graphs were finally selected for quantitative metrics analysis. The root mean square error was reduced by 4.2%-73.9%, the mean absolute error by 7.1%-93.4%, the mean percentage error by 7.7%-86.7% and the coefficient of determination by 1.3%-134.6%. The hybrid prediction model not only avoided the pollution of water resources by the cobalt ion concentration discharged in the purification, which is also of practical significance for the technicians to control the input quantity of zinc powder according to the prediction data in time and reduce the waste of resources.

Printed polylactic acid/akermanite composite scaffolds for bone tissue engineering; development and surface modification

The susceptibility of bone tissues to various factors such as ageing, accidents, and diseases has led to extensive tissue engineering research focusing on bone tissues. Hence, this research also aims to determine the optimal amount of Akermanite (AK) addition to the polylactic acid scaffold for bone tissue engineering applications, as well as the effects of surface modification on its properties. The Akermanite was synthesized using the sol-gel method. Then, composite scaffolds of polylactic acid, including 0, 10, 20, and 30 wt% AK, were printed via the fused deposition modelling (FDM) process. These scaffolds were labelled as PLA, 10 wt% AK, 20 wt% AK, and 30 wt% AK, respectively. The X-ray diffraction analysis confirmed the production of the AK high-purity phase. Cell viability tests on composite scaffolds confirmed non-toxicity, and cell adhesion improved with AK addition. Mechanical testing showed that the compressive strength of composite scaffolds increased by increasing the AK content of the composite. This study recommended the 20 wt% AK scaffold as the optimal composition for bone tissue engineering. The surface-modification of polylactic acid/AK composite scaffolds using sodium hydroxide showed that it can be suitable for advanced tissue structures and medical applications, contributing to advancements in tissue engineering and medical technology for improved bone treatments.

An optimization method for terahertz metamaterial absorber based on MOPSO

Metamaterials can freely control terahertz waves to obtain the desired electromagnetic characteristics by designing the geometry and direction of the unit structure, which is widely used in sensing, communication and stealth technology in radar. The traditional design of terahertz metamaterial absorber usually requires continuous structural adjustment and a large number of simulations to meet the expected requirements. The process is heavily dependent on the experience of researchers, and the physical modeling and simulation solution process is time-consuming and inefficient, which has greatly hindered the development of metamaterial absorbers. Therefore, deep learning has been used to predict the structural parameters or spectra of metamaterial absorbers due to its powerful learning ability. However, when designing a new structure, a large number of training samples need to be reprepared, which is time-consuming and not universal. Particle swarm optimization can quickly converge to the optimal solution through the sharing and cooperation of individual information in the group, without prior preparation. Therefore, this paper proposed a fast design method of terahertz metamaterial absorber based on multi-objective particle swarm optimization algorithm. Taking a new center symmetric absorber structure composed of four L as an example, the structure parameters are optimized to achieve fast automatic design of metamaterial absorber. The multi-objective particle swarm optimization takes the absorptivity and quality factor as independent targets to design the structure parameters of the absorber, realizes the dual-objective optimization of the absorber, and overcomes the shortcoming of the multi-objective conflict that PSO cannot solve. The optimally-designed absorber achieves perfect absorption at 1.613THz with a quality factor of up to 319.72 and a sensing sensitivity of 264.5GHz/RIU when used for refractive index sensing. In addition, the causes of absorption peaks are analyzed in detail using impedance matching, surface current, and electric field distribution. By studying the polarization characteristics of the absorber, it is found that it is not sensitive to polarization, which is more stable in practical application. In summary, the multi-objective particle swarm optimization algorithm can realize the design according to the demand, reduce the experience requirement of researchers in the design of metamaterial absorber, improve the design efficiency and performance, and have great application potential in the design of terahertz functional devices.

A method for determining stability range of grid‐side converter parameters of PMSG based on negative feedback modelling of phase‐locked loop and outer control loop

AbstractDirect‐drive wind farms are integrated into power grid through power electronic devices, extensively utilizing voltage source converters (VSCs) for grid connection, which has precipitated frequent stability challenges within the grid integration systems of wind farms; it is imperative to properly tune the VSC parameters consequently. In grid‐connected VSC system, the interaction between phase‐locked loop (PLL) and outer control loop (OCL) has a significant impact on system stability. Therefore, it is essential to account for the interactions when establishing the stability boundaries for VSC parameters. This article proposes a new method to determine the parameter stability range in grid‐side converter of direct‐drive permanent magnet synchronous generator (PMSG), based on the interactions between PLL and OCL. This method simplifies the modelling process and improves the accuracy of stability range calculations compared to traditional methods. The system is modelled as a negative feedback system with PLL as the forward path and OCL as the feedback path. Both paths include grid dynamics. This modelling approach not only captures the interaction between PLL and OCL, but is also straightforward and clear. Using the Bode diagram, stability mechanism of the system considering interaction between PLL and OCL is studied. The VSC parameter stability criterion of the system is analysed based on the open‐loop frequency characteristics of the system. Then an analytical description of the stability range for parameters is provided. Simulation results confirm the effectiveness of our proposed method.