Essential oils can be used as an alternative to using synthetic insecticides for pest management. Therefore, the insectistatic and insecticidal activities of the essential oil of aerial parts of Salvia ballotiflora (Lamiaceae) were tested against the fall armyworm Spodoptera frugiperda (Lepidoptera: Noctuidae). The results demonstrated insecticidal and insectistatical activities against this insect pest with concentrations at 80 µg·mL⁻¹ resulting in 20% larval viability and 10% pupal viability. The larval viability fifty (LV₅₀) corresponded to a concentration of 128.8 µg·mL⁻¹. This oil also increased the duration of the larval phase by 5.5 days and reduced the pupal weight by 29.2% withrespect to the control. The GC-MS analysis of the essential oil of S. ballotiflora showed its main components to be caryophyllene oxide (15.97%), and β-caryophyllene (12.74%), which showed insecticidal and insectistatical activities against S. frugiperda. The insecticidal activity of β-caryophyllene began at 80 µg·mL⁻¹, giving a larval viability of 25% and viability pupal of 20%. The insectistatic activity also started at 80 µg·mL⁻¹ reducing the pupal weight by 22.1% with respect to control. Caryophyllene oxide showed insecticidal activity at 80 µg·mL⁻¹ giving a larval viability of 35% and viability pupal of 20%.The insectistatic activity started at 400 µg·mL⁻¹ and increased the larval phase by 8.8% days with respect to control. The LV₅₀ values for these compounds were 153.1 and 146.5 µg·mL⁻¹, respectively.

Early leak detection in water distribution networks is essential to minimize losses and improve operational efficiency. This systematic review analyzes 53 studies published between 2018 and 2025 that employed machine learning, deep learning, and hybrid approaches. The results show that pressure is the most widely used and most sensitive input variable for identifying hydraulic anomalies. Most datasets originate from EPANET-generated simulations, while experimental and field data are less common due to their high costs and operational complexity. Machine learning models, particularly SVMs, achieve accuracies between 94 and 100%, demonstrating stability with noisy data and low computational cost, while in deep learning, CNNs are most effective for multiclass classification and localization, typically reaching 95–99% accuracy. Hybrid approaches that combine automatic feature extraction (e.g., CNNs or autoencoders) with conventional classifiers (such as SVMs or LSSVMs) yield the best results, surpassing 97% accuracy and achieving localization errors below 0.2 m. Based on these findings, a theoretical model is proposed using a hybrid CNN + SVM approach to enhance accuracy, robustness, and adaptability in real-time monitoring systems.