Intelligent Decision Technologies

This study proposes an integrated framework combining biomechanics, spatial econometrics, and machine learning to enhance sustainable livestock management. Using panel data from 11 provinces over a two-decade period, we analyse the nonlinear relationship between livestock concentration and regional economic output using Cobb-Douglas functions and dynamic panel regression with GMM estimation. Location entropy indices reveal spatial disparities in livestock agglomeration, while robustness is ensured via VIF, root tests, and LSDVC correction. A case study on meerkat behavior demonstrates the effectiveness of a hybrid SVM model in classifying key behaviors—vigilance, resting, foraging, and running—from over 82,000 labeled video bouts. The model achieves high classification accuracy and interpretability. Results highlight the synergy of biomechanical analysis and intelligent classification in improving animal welfare monitoring, while spatial models inform balanced regional livestock development. The proposed framework offers actionable insights for eco-agriculture policy, smart farming systems, and sustainable rural economic planning.

This study examines the cost pass-through behavior in China's retail electricity market by comparing vertically integrated and independent retailers. Using a unique monthly panel dataset covering 34 provincial regions and 340 retail pricing plans) 10 retailers × 34 province) from 2018 to 2023, the study disaggregates upstream electricity costs into generation costs proxied by futures market prices and regulated transmission and distribution charges (“lines costs”). The empirical analysis, based on fixed effects panel regressions, reveals that independent retailers exhibit significantly higher pass-through rates, particularly for generation-related expenses, while vertically integrated firms demonstrate pricing stability due to internal hedging strategies. The findings further indicate that lines costs are more uniformly passed through across all firm types due to their regulated and predictable nature. Transmission length, transformer size, and customer population are important infrastructure factors influencing the price of electricity. This research sheds light on the firm's structural configuration, the regulatory environment and the firms costs and how the three are interrelated to explain price behavior under mixed markets for electricity. The contribution of the study will inform policymakers on the issue of electricity market liberalization and the necessary tools to implement it. This is particularly important for developing countries that are shifting to the liberalized competitive retail model.

Printed circuit boards (PCBs) are essential components in electronic systems, providing mechanical and electrical support for various integrated elements. Due to their intricate designs, PCBs are prone to manufacturing defects such as short circuits, spurs, and unintended copper formations, which can negatively impact device functionality and reliability. This paper proposes an enhanced You Only Look Once version 8 (YOLOv8) model for detecting and localizing defects on PCBs. The proposed model integrates Coordinate Attention (CoordAtt) and a Bi-directional Feature Pyramid Network (BiFPN) into its architecture to improve detection accuracy. Additionally, it outputs annotated images and visualises the defect detection process using heatmaps to facilitate effective defect localisation. The model has been evaluated using two public benchmark datasets: DeepPCB and TDD-NET. The results of the experimental evaluations demonstrate the effectiveness of the proposed Coord-BiFPN-YOLOv8 model. It achieved on the DeepPCB dataset a mean Average Precision at an Intersection over Union (IoU) of 0.5 (mAP50) of 93.4%, and for mAP50:95 the model achieved 65.9% accuracy. Also, on TDD-NET, the proposed model achieved an impressive mAP50 of 99.1% and an mAP50:95 of 67.5%. These results highlight the model’s high precision and robustness in detecting PCB defects. Additionally, the proposed model outperforms the standard YOLOv8 in defect identification, with improvements of 1.8% in F1-score and 6.4% in mAP50:95.

This investigation aims to explore users’ accuracy in identifying computer mice designed with the golden ratio and their aesthetic preferences, with a specific focus on gender differences. First, the study used a database of 233 Genius brand mice and selected five samples with length-to-width ratios ranging from 1.42 to 1.78. A total of 99 university students (62 females and 37 males) participated in absolute and relative judgment tests. Results showed that most participants failed to correctly identify the mouse with a ratio closest to the golden ratio (approximately 1:1.62), even when provided with 2D or 3D visual references. The chi-square test showed that the difference in recognition accuracy between males (32%) and females (34%) was not statistically significant (p = 0.34). In addition, there was no correlation between aesthetic preference and whether the mouse followed the golden ratio. These findings suggest that golden ratio design is neither easily recognizable nor a key factor in product preference, and designers need not overly rely on it to enhance product appeal. Finally, polynomial regression analysis was employed to investigate the effect of the ratio on the preferences of all participants. The model fit was very high (R² = 99.91%, p-value = 0.038), showing that the mouse ratio was associated with all participants’ preferences. According to the results, if a ratio/preference prediction system for mouse design is developed in the future, a mouse design with an aspect ratio of 1.5 to 1.56 can serve as an important reference indicator for preference prediction.

With the help of a smart energy hub as well as a dynamic demand response (DR) program, this study suggests a comprehensive energy management strategy for an intelligent microgrid. The main goal is to reduce pollutants as well as total operating expenses while maintaining a reliable and adaptable energy supply in face of renewable energy uncertainty. A number of dispersed energy sources, including solar panels, WTs, micro turbines, solid oxide fuel cells, CHP units, as well as battery energy storage devices, are integrated into the suggested system. Demand-side flexibility is achieved through the integration of real-time DR systems and home charging for electric vehicles. Every component is represented using detailed physical and operational constraints. The smart energy hub serves as a central coordinator for balancing energy generation, conversion, storage and consumption, while facilitating market participation. Utilizing scenario-based modeling of solar and wind variability, the Non-dominated Sorting Genetic Algorithm II is employed to create and solve a nonlinear multi-objective optimization problem. The method is evaluated using two case studies, one with DR and one without. The findings demonstrate that incorporating DR significantly lowers emissions and operational costs by 9%, while increasing renewable energy production by up to 15%. Battery scheduling effectively correlates with price changes, whilst smart energy source dispatch decreases reliance on grid imports. Under high renewable integration, the results show that the proposed coordinated strategy improves microgrid flexibility, sustainability, and economic performance.

Wireless contactless sensing technologies have become vital for individualized healthcare monitoring since they let doctors check on patient’s health without having to touch them. However, it is still hard to choose the best sensing technology because of uncertainty, imprecision, and different expert perspectives. For the sake of solving such problems, a complete decision-making framework based on the Neutrosophic Soft Fuzzy TOPSIS method has been proposed which works well with vague and missing information in evaluations with more than one criterion. The model systematically evaluates different sensing technologies based on important criteria and gives a clear and strong score by calculating how near they are to the perfect solution. A comprehensive discussion and sensitivity analysis on the results obtained have also been presented along with pros and cons. A table comparing the proposed model to other decision-making methodologies has been additionally provided keeping the benchmarking into account to ensure the model is both accurate and consistent in its performance. Also, comparative trials show that the proposed method is better at coping with uncertainty and competing expert opinions. Overall, in this framework, results show that the proposed decision-support tool is dependable and easy-to-understand for healthcare professionals and policymakers. It helps them choose the best contactless human sensing technologies for real-world personalized healthcare applications.

With the rapid development of global navigation satellite system (GNSS) technology, IGS signals play a crucial role in many fields such as positioning, navigation, and time synchronization. Nevertheless, the multipath error problem seriously affects the performance of IGS signals in practical applications. In this paper, the multipath error problem of IGS signals is studied in depth, and an improvement method based on a digital filtering technique is proposed. The article first provides a comprehensive analysis of the impact of multipath error and identifies the shortcomings of existing studies, thus clarifying the motivation of this study. Subsequently, this paper uses digital filtering techniques to estimate the multipath error accurately and designs a corresponding improvement strategy. Through experimental verification, this paper demonstrates the significant effect of the improved method in improving signal quality and positioning accuracy. Finally, the article summarises the research results. It discusses their potential promotion value in IGS signal reception and related applications, which provides important theoretical support and practical guidance for future technology development and applications.

In response to the difficulties in integrating multimodal data and insufficient model generalization ability in traditional cross-modal knowledge transfer, this article used the Transformer model to explore it in the new generation learning space. Firstly, the article analyzed the processing methods of data and models in cross-modal knowledge transfer, and explored the application of Transformer models in the learning space. This model used natural language processing to represent and extract textual features, Mel Frequency Cepstral Coefficients (MFCCs) to represent and extract audio features, and Faster R-CNN (Faster Region-based Convolutional Neural Network) to represent and extract image features. The article also discussed the implementation process of the Transformer model functionality. The experiment used data from four datasets, including Quora Question Pairs, to test the performance of the model's cross-modal knowledge transfer through intelligent question answering and task analysis. In single type data testing, the accuracy and recall of the model in this article were better than the comparison model in the three types of data. The highest accuracy and recall in the test set were 91% and 93%, respectively. In the most challenging multimodal intelligent question answering test, the speech-image question answering method achieved an accuracy rate of 89% in answering open questions, indicating that the model had good multimodal data fusion ability. In the analysis experiment of 6 homework prone knowledge points on images with text annotations, the induction accuracy reached 85%, indicating that the model had strong generalization ability. The experimental results showed that the Transformer model had good cross-modal knowledge transfer performance, providing a reference for subsequent research on cross-modal knowledge transfer in the new generation learning space.

Industry 4.0 is reshaping conventional factories into “smart factories” via the widespread use of IoT-enabled networks of linked devices, sensors, and software for process optimization and monitoring. Intelligent manufacturing facilities may employ IoT-based predictive maintenance to reduce downtime, increase equipment longevity, and avoid machine problems. Manufacturers may get real-time insights into energy consumption patterns, which is a major concern in the business. The primary objective is to optimize energy use during part manufacturing. Hence, this paper proposes the Internet of Things- Low-Power Wide-Area Network Model (IoT-LPWM) to monitor manufacturing and reduce energy consumption. The proposed method's production status component uses visual Knowledge Map Analysis loaded with data from the edge device. A Low-power wide-area network (LPWAN) is the fundamental component of the suggested approach to industrial wireless communication. Using edge computing technology in LPWAN helps reduce computational complexity by shifting high-intensity processing to the periphery, where devices with computing resources are more readily available. Both the energy needed to process and store massive data and the likelihood of cyberattacks may be decreased with this method. The experimental results show that the IoT-LPWM provides useful information to help them make decisions and reduce energy consumption. The experimental results show that our proposed method IoT-LPWM achieves a high performance ratio of 97%, attack prevention ratio of 96.3%, energy management ratio of 93.8%, and data transmission ratio of 98.1% compared to other methods.

Warehouse automation relies on accurate item sorting to boost production while cutting workforce needs. Intelligent warehouse item sorting robots have become indispensable for job scheduling, particularly for the best distribution and order of sorting jobs. Optimizing job scheduling, robot usage, and coordination at the centralised control system level can maximise warehouse sorting robots’ productivity by allocating tasks more effectively. The paper proposes an Enhanced Chimpanzee Optimization algorithm-based Task Scheduling Framework (ECOA-TSF) to meet warehouse automation requirements. The proposed method utilizes communication, grouping, and learning techniques for warehouse item sorting. The ECOA-TSF is a local search mechanism to improve job sequencing, allocation, and optimization of item sorting operations, making it more adept at exploitation and exploration. The efficacy of the ECOA-TSF proposed technique is evaluated via comprehensive simulation results compared to more conventional scheduling methods. Based on the findings, the ECOA-TSF outperforms the baseline techniques regarding work completion time, resource utilization, and overall effectiveness of the system. This work provides a trustworthy and efficient task-scheduling method for warehouse product sorting.

Analyzing customer opinions is one of the important approaches for business development. Considering the expansion and diversity of airlines, an accurate understanding of passengers’ opinions can be an effective step in surpassing companies in the competitive market. Therefore, a model for categorizing airline customers’ thoughts utilizing the Twitter platform was attempted to be presented in this study. This study's suggested model is predicated on two primary methods: Neural networks and text analysis. For this purpose, two layers of CNN and DNN were combined as the main layers in three different dimensions, and six models were formed. Then, by carrying out a case study and looking at several assessment indices, it was discovered that the models’ accuracy declined as their dimension rose. In other words, one-dimensional models are more accurate than two- and three-dimensional models in classifying textual data. On the other hand, among one-dimensional models, the CNN-based model has less error. Therefore, the one-dimensional CNN model was suggested as a proposed model to classify the opinions of airline passengers.

In response to the difficulties in integrating multimodal data and insufficient model generalization ability in traditional cross-modal knowledge transfer, this article used the Transformer model to explore it in the new generation learning space. Firstly, the article analyzed the processing methods of data and models in cross-modal knowledge transfer, and explored the application of Transformer models in the learning space. This model used natural language processing to represent and extract textual features, Mel Frequency Cepstral Coefficients (MFCCs) to represent and extract audio features, and Faster R-CNN (Faster Region-based Convolutional Neural Network) to represent and extract image features. The article also discussed the implementation process of the Transformer model functionality. The experiment used data from four datasets, including Quora Question Pairs, to test the performance of the model’s cross-modal knowledge transfer through intelligent question answering and task analysis. In single type data testing, the accuracy and recall of the model in this article were better than the comparison model in the three types of data. The highest accuracy and recall in the test set were 91% and 93%, respectively. In the most challenging multimodal intelligent question answering test, the speech-image question answering method achieved an accuracy rate of 89% in answering open questions, indicating that the model had good multimodal data fusion ability. In the analysis experiment of 6 homework prone knowledge points on images with text annotations, the induction accuracy reached 85%, indicating that the model had strong generalization ability. The experimental results showed that the Transformer model had good cross-modal knowledge transfer performance, providing a reference for subsequent research on cross-modal knowledge transfer in the new generation learning space.

With the emergence of e-commerce, more and more people conduct transactions through the Internet, thus resulting in a large number of transaction data. Data mining is to decompose a large amount of data according to data rules, and analyze network transaction data, so as to provide necessary digital links for companies to analyze the market and develop business. Although time series data mining is smaller than other types of data mining, it is also an important issue of data mining. In the real world, the correlation between data and time is very common. The study of time series model plays a very important role in data mining. Due to different purposes, Taobao data analysis is also different. In addition to statistics, at present, the in-depth research and analysis of Taobao data are relatively insufficient, and the analysis of Taobao transaction data based on time series is rare. In order to improve the accuracy of Taobao transaction data mining and better formulate Taobao marketing strategy, this paper used time series data mining technology to mine Taobao transaction data. This paper first introduced the role of Taobao transaction data mining, and then described the calculation method of time series data mining, including the re-description of time series and the similarity measurement of time series. Finally, through a series of processes such as data collection, data processing and data feature extraction, the data mining model for Taobao transaction was established, and two data prediction evaluation indicators, namely prediction accuracy and entropy, were proposed. The experimental part verified the effect of Taobao transaction data mining. The experimental results showed that the data mining model moment had good data prediction accuracy and entropy. The average data prediction accuracy was 94.26%, and the data mining ability was strong.