Customer churn is a key concern in numerous companies, including the telecom industry. Decision-makers and business analysts feel that retaining existing consumers is less expensive than acquiring new clients. To provide a retention solution, customer relationship management (CRM) analysts must recognize customers who intend to quit the company and understand patterns of behavior from existing churn customers' data. This paper provides a Hybrid Prediction Model (HPM) that employs classification and clustering approaches to predict client attrition. To pick the key characteristics, the proposed model uses an RFE algorithm, filter method, Boruta algorithm, and correlation matrix, as well as Density-based Spatial Clustering of Applications with Noise (DBSCAN) to discover and eliminate outlier data. In addition, it uses stacking classifier to categorize consumers, tuning threshold to handle data imbalance, and k-mean algorithm to segment churning customers—whom the stacking classifier has classified—into groups to make group-based retention offers. The Telecom dataset is not publicly available due to protect the private information of customers. Thus, the data used in this study was obtained from Openml Website. The data set contains 5000 observations and 21 variables. Moreover, the proposed model can be easily adapted and applied on larger datasets. Several metrics are used to evaluate the proposed hybrid prediction model, including accuracy, recall, precision, receiving operating characteristics (ROC) area, and f-measure. The results show that our hybrid model outperforms single techniques. Furthermore, when changing the threshold in terms of recall metrics, the model performs better.

This study investigates the use of high-strength reinforcements (ASTM A1035 Grade 100) instead of conventional reinforcement (such as ASTM A615 Grade 60) in a blast-resistant reinforced concrete flexural member. The stress-strain curve of an ASTM A1035 Grade 100 steel differs from conventional grade 60 steel. Grade 100 steel possesses higher strength and an unclear yield threshold. When designing a flexural member to resist blast loading, its resistance function is typically developed utilizing the Unified Facilities Criteria (UFC 3-340-02) manual. The UFC approach simplifies the behavior of steel reinforcement to an elastic portion up to the yield point, followed by a perfectly plastic plateau. With the absence of a clear yield point in high-strength steel, this simplification may not be well represented. The dynamic performance for flexural concrete members reinforced with high-strength steels is developed using the generalized single-degree-of-freedom (SDOF) system. The SDOF performance generated with the simplified approach is compared with a more realistic resistance function of the system obtained by a computational model established with the virtual work principle and the moment curvature of a section. The results indicated that using the UFC approach for the high-strength steel flexural member may require an adjustment factor in order to reach conservative predictions when the reinforcement ratio is between 0.4% and 2%. The results also illustrated a 50% reduction in the area of steel reinforcement when using grade 100 instead of grade 60, indicating less steel congestion in the high-strength reinforced cross-section.

Renewable energy planning is set to be transformed by the integration of advanced solar irradiance forecasting techniques. By harnessing the predictive power of Machine Learning (ML) algorithms, planners can gain more accurate insights into future solar irradiance levels. This study investigates the use of ML algorithms for solar irradiance forecasting, intending to enhance planning strategies for renewable energy sources (RES) in Saudi Arabia. Using datasets sourced from various regions in Saudi Arabia, several regression models are evaluated, including Gradient Boosting Regressor (GBR), Linear Regression (LR), Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (Bi-LSTM), and K-Nearest Neighbor (KNN). The analysis of this research reveals that ensemble techniques such as Random Forest Regression (RFR) and data-driven approaches like KNN exhibit superior performance compared to conventional regression models like LR, underscoring the significance of various ML methods in solar irradiance prediction When compared to Decision Tree Regressor (DTR) and RFR, models with high goodness of-fit metrics (R-squared, adjusted R-squared) and low error metrics (Mean Absolute Error (MAE), Root Mean Square Error (RMSE)) show better predictive power. The precision with which the proposed models forecast solar irradiance levels is further confirmed by comparison with previous studies. Planning for RES is advanced by this study’s identification of the best ML methods for predicting solar irradiation. The results highlight the potential of using ML approaches to optimize solar energy system integration and accelerate the shift to sustainable energy practices.