To accurately estimate the traffic states of electric two-wheeler (ETW) flow and quantify its uncertainty, while also exploring the relationships between traffic entropy and traffic states, this paper first classifies the ETW traffic states into free, constant, and congested states using the fuzzy C-means (FCM) clustering algorithm, based on multiple traffic flow parameters as features. Subsequently, multiple traffic entropy indicators are constructed based on information entropy theory to characterise the uncertainty of ETW flow. Finally, the traffic state estimation results are used as labels, and explainable machine learning is employed to quantify the nonlinear relationships between entropy and states. The results indicate that congestion represents a relatively orderly state with lower uncertainty, while the free and constant states are relatively chaotic with greater uncertainty. This research not only provides a theoretical foundation for ETW flow uncertainty but also introduces a novel approach for estimating traffic states in non-motorised traffic flow.

Bridge deck deterioration poses a significant challenge to transportation infrastructure, resulting in costly maintenance and potential safety hazards. Traditional bridge deck assessments primarily rely on visual inspections, which can be subjective and fail to capture subsurface defects, such as delamination, rebar corrosion, and concrete degradation. To enhance the accuracy of condition assessment, this study explores multi-sensor data fusion and clustering techniques for defect identification using Ground Penetrating Radar (GPR) and Impact Echo (IE). By integrating multiple Non-Destructive Evaluation (NDE) datasets, a clustering-based framework was developed to automatically categorize bridge deck conditions. K-Means, Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Gaussian Mixture Models (GMM), and Fuzzy C-Means (FCM) clustering algorithms were evaluated to determine their effectiveness in grouping similar defect patterns. The optimal number of clusters is determined using the Elbow Method, Silhouette Score, and Davies-Bouldin Index. Results indicate that DBSCAN outperforms other clustering techniques in detecting defect hotspots while effectively handling noise and spatial inconsistencies. The clustered defects are mapped spatially to visualize regions of deterioration, enabling bridge engineers to identify high-risk areas and prioritize maintenance efficiently.

Türkiye is highly susceptible to earthquakes due to its active tectonic structure and the presence of major fault lines. The accurate estimation of earthquake magnitudes is essential for effective risk mitigation and structural resilience. This study proposes an integrated methodology combining clustering, statistical modeling, and deep learning techniques for the analysis and forecasting of earthquake magnitudes. Initially, earthquakes are classified into three distinct regions using the Fuzzy C-Means (FCM) clustering algorithm. For each region, statistical distributions are applied to characterize magnitude behavior. Subsequently, the Long Short-Term Memory (LSTM) model is used to predict future earthquake magnitudes. The joint application of these three methods provides a comprehensive framework for regional seismic analysis. The findings suggest that the Gumbel distribution offers the best fit for modeling return periods of earthquakes with magnitudes greater than Mwg 5.18, where Mwg denotes the global moment magnitude. Estimated return periods range from 2.56 to 3.63years in the first region, 2.53 to 3.55years in the second region, and 2.71-4.22years in the third region, based on probability levels between 25 and 95%. The LSTM model forecasts that the third region is likely to experience relatively stronger seismic activity, with maximum magnitudes ranging from 2.4 to 6.5 between October 2021 and March 2029. For the same period, expected magnitudes in the first and second regions range from 2.0 to 5.7. These forecasts are supported by model performance metrics that confirm the projected magnitudes are within an acceptable and reliable range of accuracy for medium-term seismic forecasting.

Efficient drug distribution is fundamental to ensuring the quality of public healthcare services. However, health departments often face challenges with imbalances between drug demand and available supply. This study addresses this issue by applying the Fuzzy C-Means (FCM) clustering algorithm to categorize drug demand levels across 16 public health centers (puskesmas) in Langkat Regency, Indonesia, from 2021 to 2023. Using historical data from 2,400 drug records, the analysis identified five distinct demand clusters: Very Low, Low, Medium, High, and Very High. The results revealed a significant disparity in drug needs, with the "Very High" demand cluster dominating (51.29% of data) in centers like Besitang and Tanjung Selamat, driven by high morbidity rates. In contrast, other clusters were less prevalent, such as the "Low" demand cluster, which was primarily concentrated in the Gebang health center. These findings, visualized using t-SNE plots, highlight significant regional variations in pharmaceutical needs. This data-driven clustering provides a robust framework for the Langkat District Health Office to develop more targeted, efficient, and equitable drug distribution strategies, ultimately improving healthcare service delivery.

Customer segmentation allows companies to create mutual profiles of their customers. Determining industrial customer segments based on a single perspective causes various customer features to be disregarded. This study aims to develop a holistic segmentation approach in a B2B setting. The paper proposes a multi-dimensional segmentation model with four main criteria: customer purchasing performance, customer cooperation, customer workload, and customer potential. The case study demonstrates the real-life application of the proposed model using 379 customer data and 17 sub-criteria under four dimensions. The Fuzzy C-Means Clustering Algorithm creates the customer segments, and the Fuzzy Analytical Hierarchical Process is used to calculate criteria weights. The marketing strategies of each segment are used to guide customer relations and managerial decisions. This paper suggests that companies segment their customers by considering financial performance, cooperation level, future potential throughput, and challenges. It provides a practical and holistic insight into industrial customer segmentation.

ABSTRACT The urban landscape environment is a dynamic system, and its imagery is strongly influenced by public perception, making quantitative measurement challenging. Traditional methods rely on real-scene virtual technology to assess preferences, but lack clear evaluation indicators, leading to subjectivity. This study proposes using computer vision colour quantisation to objectively measure urban landscape imagery and clarify public preference orientations. The fuzzy c-means (FCM) clustering algorithm is applied to pixel points in urban landscape images, producing colour quantisation and constructing a chromaticity matrix to extract local colour features. These features are then input into an extended fully convolutional network to identify landscape elements. Based on this, a quantitative measurement model linking landscape elements and imagery is developed, enabling the analysis of public preferences. Colour, as an intuitive and objective visual element, provides a strong basis for quantification. Case analyses demonstrate the method’s effectiveness in measuring urban landscape imagery, revealing differences in public preferences across landscape composition, proportion, and colour. This approach reduces subjectivity and offers a robust framework for quantitative evaluation of urban landscape environments.

Air pollution remains a critical threat to human health, leading to respiratory issues and fatalities worldwide. With industrialization and urbanization on the rise, monitoring and predicting Air Quality Index (AQI) levels, especially in cities, becomes increasingly challenging. Cluster analysis emerges as a crucial tool for discerning patterns in air pollutant data. This study introduces the Common Mahalanobis Distance-based Fuzzy C-Means algorithm (CM-FCM) for air pollution data analysis, evaluating its effectiveness against k-means Clustering Algorithm and Euclidean Fuzzy C-Means Clustering Algorithm in terms of accuracy. The CM-FCM algorithm identified non-spherical clusters that accurately captured pollution patterns, enabling precise hotspot detection. Applied to datasets from Byrnihat and Indira Gandhi International Airport (LGBI) Airport, India, it categorized LGBI Airport as “Moderately Polluted” and Byrnihat as “Most Polluted”, with PM10 levels exceeding World Health Organization (WHO) standards on 99% of days. By considering pollutant correlations, CM-FCM provides valuable tools for policymakers to address pollution hotspots and enhance public health strategies.

This paper proposes a novel ensemble method with multiple runs of the fuzzy c-means (FCM) clustering algorithm as the base learners. However, the major challenge in the ensemble of different runs of FCM lies in the labeling of clusters in each run. As there is no uniformity for numbering the cluster labels, the data points belonging to close clusters may be assigned to different clusters across multiple runs of FCM. Hence, assembling the cluster solutions of different runs of FCM becomes difficult. This challenge has been addressed in this study by proposing a novel method to ensemble multiple runs of FCM. In the proposed method, the cluster labels of different runs are renumbered based on proximity among the clusters, making it easy to ensemble cluster solutions in multiple runs of FCM. The class labels of the non-ensemble data points are determined by using three state-of-the-art classifiers, namely, k-nearest neighbors, support vector machine, and artificial neural network. The proposed method is applied for segmenting magnetic resonance imaging (MRI) of the human brain. The division of MRI brain images into distinct tissue classes is a crucial aspect of neurological disease and clinical research. The performance of the proposed method in segmenting the MRI brain images is compared to the Gaussian mixture models based on the values of three performance measures, namely, the Rand index, adjusted Rand index, and Minkowski score. The simulation results demonstrate the supremacy of the proposed method over the Gaussian mixture models.

Photovoltaic (PV) power generation forecasting is crucial for grid stability, but existing models perform poorly during abrupt weather changes or significant data fluctuations. This paper proposes a method to improve prediction accuracy through weather classification, signal decomposition, and parameter optimization. First, the correlation between meteorological indicators and power generation is analyzed. The Pearson correlation coefficient method is used to screen out the main meteorological factors, and the Fuzzy C-Means (FCM) clustering algorithm is applied to classify weather types. For data from different weather categories, Variational Mode Decomposition (VMD) is used to process the original signal, separating unstable and fluctuating components. Subsequently, the Whale Optimization Algorithm (WOA) optimizes the parameters of the Long Short-Term Memory (LSTM) network, improving the model’s prediction accuracy. Finally, comparative analysis of the prediction results under different weather types and seasons demonstrates that the proposed method significantly reduces the Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) compared to the traditional LSTM model in various weather and seasonal conditions, resulting in higher prediction accuracy and better stability. The research results provide a reference for the daily operation and maintenance of PV power plants and offer new solutions for time series prediction problems.

To solve the problem of high cost and low efficiency of measuring equipment in traditional distribution network fault location, a fault section location and line selection strategy combining dynamic binary particle swarm optimization (DBPSO) configuration and fuzzy C-means (FCM) clustering is proposed in this paper. Firstly, the DBPSO algorithm is used to optimize the configuration scheme of the distributed voltage and current sensing device, which reduces the number of measuring devices and system cost on the premise of ensuring the global observability of the distribution network. When a fault occurs in the distribution network, the sensor device based on optimal configuration collects fault feature data, combines it with the FCM clustering algorithm to classify nodes according to fault feature similarity, and divides the most significant fault-affected section as the core fault area. Further, by calculating the Euclidean distance between each node in the fault section and the cluster center, the fault line is accurately identified. Finally, a fault simulation model based on an IEEE 11-node system is constructed to verify the effectiveness of the proposed method. The results show that, compared with the traditional fault section location and route selection strategy, this method can reduce the number of measurement devices optimally configured by 19–36% and significantly reduce the number of algorithm iterations. In addition, it can realize rapid fault location and precise line screening at a low equipment cost under multiple fault types and different fault locations, which significantly improves fault location accuracy while reducing economic investment.

Unsupervised feature selection via fuzzy c-means clustering and binary atom search algorithm

Accurate prediction of the Normalized Difference Vegetation Index (NDVI) is crucial for regional ecological management and precision decision-making. Existing methodologies often rely on smoothed NDVI data as ground truth, overlooking uncertainties inherent in data acquisition and processing. Fuzzy time series (FTS) prediction models based on the Fuzzy C-Means (FCM) clustering algorithm address some of these uncertainties by enabling soft partitioning through membership functions. However, the method remains limited by its reliance on expert experience in setting fuzzy parameters, which introduces uncertainty in the definition of fuzzy intervals and negatively affects prediction performance. To overcome these limitations, this study enhances the interval type-2 fuzzy clustering time series (IT2-FCM-FTS) model by developing a pixel-level time series forecasting framework, optimizing fuzzy interval divisions, and extending the model from unidimensional to spatial time series forecasting. Experimental results from 2021 to 2023 demonstrate that the proposed model outperforms both the Autoregressive Integrated Moving Average (ARIMA) and conventional FCM-FTS models, achieving the lowest RMSE (0.0624), MAE (0.0437), and SEM (0.000209) in 2021. Predictive analysis indicates a general ecological improvement in the Aksu region (Xinjiang, China), with persistent growth areas comprising 61.12% of the total and persistent decline areas accounting for 2.6%. In conclusion, this study presents an improved fuzzy model for NDVI time series prediction, providing valuable insights into regional desertification prevention and ecological strategy formulation.

Driving risks assessment and in-vehicle warning design for improving work zone safety.

As the grid-connected capacity of distributed photovoltaic (PV), energy storage, electric vehicle (EV), and other devices gradually increases, new source-load equipment becomes an important demand response (DR) resource in the distribution network (DN). To fully utilize the DR's capability for EVs and other devices and reduce the system operating costs and line network loss, this article presents a DR scheduling strategy for EVs based on a time-of-use (TOU) price dynamic adjustment mechanism. First, a fuzzy C-mean (FCM) clustering algorithm is used to calculate the typical operating curves of PV and electrical load and their optimal number of classifications. The deterministic scenarios express the PV's output characteristics and the users' electricity consumption characteristics. Second, a dynamic adjustment mechanism of TOU price is proposed based on the load operation curve of the DN, and the interactive price-incentive signal for DR within the DN is formulated. Finally, a DR scheduling strategy for EVs in the DN that considers the economic cost of system operation and line network loss is proposed. CPLEX in MATLAB is employed to simulate the cases. After applying the TOU price dynamic adjustment mechanism proposed, the peak total load and peak–valley load difference decreased by 6.9% and 33.8%, respectively, compared to implementing fixed electricity prices. At the same time, the operating revenue of the distribution network increased by 13.2%, and the line network loss decreased by 12.9%. The analysis results demonstrate that the proposed EV DR scheduling strategy can realize the price guidance and orderly scheduling of EVs and reduce the operation cost and line network loss in the DN.

Heterogeneous Sensor Networks (HSNs) have emerged as a prominent and versatile paradigm for various real-world applications, encompassing diverse sensor nodes with varying capabilities and characteristics. The integration of heterogeneous devices with distinct energy levels, computing power, and communication ranges poses unique challenges and opportunities in the design and operation of such networks. This paper presents a novel approach that leverages the Fuzzy C-means (FCM) clustering algorithm for cluster formation and the Modified Grey Wolf Optimization (MGWO) algorithm for Cluster Head (CH) selection in heterogeneous sensor networks. The primary goal of this approach is to optimize node energy consumption and extend the overall network lifespan. The FCM algorithm accommodates the heterogeneity of sensor nodes by allowing data points to belong to multiple clusters with varying degrees of membership. Subsequently, the Modified GWO algorithm, an updated version of GWO, incorporates Levy flight and decay function to balance exploration and exploitation phases effectively. Additionally, we enhance the MGWO fitness function by incorporating four parameters: residual energy, communication distance, connection requests to the node, and maximum communication region. Results revealed that proposed model is outperforming traditional MOD- LEACH, MGEAR and TEZEM models in terms of dead nodes, stability period, instability period and network lifetime to prove its supremacy.

The study focused on analyzing shoreline changes along the western beaches of Mersin Province, located on Turkey’s Mediterranean coast. Landsat satellite imagery from 1985 to 2022 was used to detect long-term coastal alterations. The Google Earth Engine (GEE) platform facilitated data acquisition, classification, and edge detection. A Support Vector Machine (SVM) classification algorithm was applied to distinguish land from water. To enhance classification accuracy, additional indices—Normalized Difference Water Index (NDWI), Modified NDWI (MNDWI), and Normalized Difference Moisture Index (NDMI)—were incorporated alongside Landsat spectral bands. The Canny edge detection algorithm was employed to delineate shorelines from the classified images. Resulting shoreline positions were analyzed using the DSAS, an open-source ArcGIS extension, to quantify erosion and accretion. Key shoreline change metrics— Net Shoreline Movement (NSM), Shoreline Change Envelope (SCE), End Point Rate (EPR), and Linear Regression Rate (LRR) —were derived from DSAS outputs. Over the 38-year study period, maximum shoreline advancement reached 588.59 meters, while maximum retreat was −130.63 meters. The highest erosion rates were −3.53 m/year (EPR) and −2.8 m/year (LRR), whereas the most pronounced accretion rates were 15.91 m/year (EPR) and 15.47 m/year (LRR). To identify spatial patterns in shoreline change, the Fuzzy C-Means (FCM) clustering algorithm was applied using the NSM, SCE, EPR, and LRR metrics. The resulting clusters were then interpreted in relation to land cover data provided by the European Space Agency (ESA) WorldCover dataset.

This study proposes a novel urban heat island detection method implemented in a GIS-based framework, designed to identify the most critical urban areas during heatwave events. The framework employs the fuzzy C-means clustering algorithm with remotely sensed land surface temperature and normalized difference vegetation index data to delineate and visualize hotspots. The proposed approach is compared with other established methods for urban heat island detection to evaluate their relative accuracy and effectiveness. This methodology integrates advanced spatial analysis with environmental indicators such as vegetation cover and permeable open spaces to assess urban vulnerability. The city of Naples, Italy, serves as a case study for testing the framework. The results from the case study indicate that the proposed method outperforms alternative methods in identifying heat hotspots, providing higher accuracy and suggesting potential adaptability to other urban contexts. This GIS-based approach not only provides a robust tool for urban climate assessment but also serves as a decision support framework that enables urban planners and policymakers to identify critical areas and prioritize interventions for climate adaptation and mitigation.

Scientific workflows and other big data applications requiring large amounts of data may be shared through cloud computing. Big data processing with scientific workflows is costly regarding bandwidth, execution time, and data transmission. To reduce these expenses, a type-2 fuzzy set-based data collaboration technique is proposed in this research. It lowers the cost of processing massive amounts of data and optimizes data placement. The suggested method examines the datasets inside each data center using data dependencies, groups the datasets using the Interval Type-2 Fuzzy C-Means (IT2FCM) clustering algorithm and then reorganizes the clusters according to data collaboration. With superior outcomes compared to prior techniques, our proposed method of using type-2 fuzzy sets to accomplish collaborative clustering may assist in dealing with data uncertainties and lower the total quantity of data placement.

Aerial image segmentation of natural disaster-impacted areas and detailed and automatic natural disaster assessment are the main focus of this study. Detecting and recognizing objects on aerial images of areas impacted by natural disasters and assessing natural disaster-impacted areas are still difficult problems. To solve these problems, this study utilizes four of the latest transformer-based semantic segmentation network models, bidirectional encoder representation from image transformers (BEIT), dense prediction transformer (DPT), OneFormer, and SegFormer, and proposes a detailed and automatic natural disaster assessment of the segmented image. The SegFormer model achieved the first-best result, and the OneFormer model achieved the second-best result. The SegFormer model outperformed OneFormer by 1.58% higher for the mean accuracy value and 4.28% for the mean intersection over union (mIoU) value. All receiver operating characteristics (ROC) curves have mean area under curve (AUC) values above 0.9, which means that the SegFormer model performs well in generating semantic segmentation images. The fuzzy c-means (FCM) clustering algorithm performed well and could automatically cluster the natural disaster assessments into four categories. This study has produced semantic segmentation of aerial images of areas impacted by natural disasters and natural disaster assessments, which can be used in natural disaster management systems.

Wireless sensor networks (WSNs) are essential for many different types of applications, including industrial, commercial, and agricultural. Inadequate node location, dynamic nodes, network longevity, increased packet drop rates, scaling problems, limited adaptability, and changing climatic conditions pose challenges to the WSN's efficacy. Numerous bio-inspired algorithms have been previously introduced for node placement, demonstrating a significant improvement in data aggregation performance. However, because of the low variability in the solution, weak convergence, and poor balance between exploitation and exploration, the results of WSNs are challenging to interpret. With a unique competitive Cuckoo search algorithm (CCSA), this research presents a connectivity-aware, energy- efficient, and node placement method. Using the elite population to increase the variety of the answer, the suggested competitive strategy aims to enhance convergence. It additionally employs the fuzzy C-mean clustering algorithm. The clustering optimization based on cluster head energy, density, location, Gini coefficients, and other comparable factors uses an artificial bee colony optimization (ABC) algorithm to enhance the clustering and cluster head position. A comparison between the recommended scheme and the traditional state-of-the-art reveals that the suggested system performs better regarding network throughput, residual energy, and network lifetime.