Universitas Darma Persada

Bencana alam tidak hanya menimbulkan kerusakan fisik, tetapi juga berdampak pada kondisi psikologis dan kesehatan anak sebagai kelompok rentan. Anak korban bencana berisiko mengalami trauma serta penurunan pemahaman terhadap perilaku hidup sehat apabila tidak mendapatkan intervensi yang tepat. Kegiatan Pengabdian kepada Masyarakat ini bertujuan untuk memberikan dukungan psikososial melalui trauma healing dan meningkatkan pengetahuan kesehatan pada anak korban bencana di Desa Teupin Banja, Kecamatan Muara Batu, Kabupaten Aceh Utara. Metode pelaksanaan meliputi tahap persiapan, pelaksanaan, evaluasi, dan monitoring dengan sasaran sebanyak 42 anak. Intervensi dilakukan melalui kegiatan bermain terapeutik dan edukasi kesehatan secara interaktif dan ramah anak. Hasil kegiatan menunjukkan adanya peningkatan partisipasi anak, ekspresi emosi yang lebih positif, serta pemahaman yang lebih baik mengenai perilaku hidup bersih dan sehat. Kesimpulannya, kegiatan trauma healing dan edukasi kesehatan mampu memberikan dampak positif terhadap pemulihan psikososial awal dan peningkatan pengetahuan kesehatan anak korban bencana.

The goal of this community service project was to improve lecturers' and teachers' abilities to use cutting-edge prompting strategies based on artificial intelligence (AI) as a breakthrough in teaching. Chain of Thought (CoT) and Role Prompting, two crucial prompting techniques that have been demonstrated to greatly enhance human-AI interaction in educational settings, were the main topics of the course. Twenty-five participants from different educational institutions participated in a series of workshops, practical exercises, and case-based discussions. The findings showed that participants' capacity to create efficient and contextually relevant prompts had significantly improved. Additionally, the training helped teachers become more technologically literate and acted as a link to assist the continuous digital transformation of education. The initiative highlights the importance of equipping educators with AI-related skills that are both practical and pedagogically meaningful, especially as generative technologies become increasingly embedded in learning environments.

Seismic magnitude forecasting in megathrust zones poses a critical challenge for disaster mitigation, necessitating a rigorous balance between prediction accuracy and computational latency. This study presents a comparative evaluation between a Transformer-based Deep Learning model (Informer) and an adaptive statistical method (Rolling ARIMA) for earthquake magnitude prediction. Case studies were conducted in two seismically active regions: Aceh (Sunda Megathrust) and Tohoku (Japan Trench), using USGS catalog data spanning 1990-2024. Experiments were performed across multiple temporal scales (daily, weekly, monthly) and prediction horizons. Results indicate that while the Informer excels in minimizing relative error (SMAPE) on large-volume datasets, Rolling ARIMA demonstrates significant superiority in shortterm accuracy ($\mathbf{R}^{\mathbf{2}}$ up to 0.92 in Aceh) and stability under limited data scenarios. Crucially, the study quantifies a massive divergence in computational efficiency: Rolling ARIMA achieves retraining in under 150 seconds with sub-second deterministic inference, whereas the Informer requires $\gt1000$ seconds for training with unstable inference latency. These findings suggest that for Edge Computing-based Early Warning Systems (EWS), adaptive statistical approaches offer a more pragmatic and reliable solution than complex deep learning architectures.

Open-pit mining in tropical regions requires continuous dewatering to ensure operational safety, yet conventional diesel-based pumping systems are energy-intensive, costly, and unsustainable. To address this challenge, this paper proposes an Internet of Things (IoT)-enabled hybrid diesel-solar photovoltaic (PV) pumping system integrated with realtime monitoring, automated control, and cloud-based communication. A rainfall-driven simulation is developed to evaluate system performance under low, medium, and extreme inflow conditions, using fieldreferenced parameters for pump load and site-specific hydrology. Monthly irradiance values of 2.1-2.9 $\mathrm{kWh} / \mathrm{m}^{2} /$ day (annual mean $2.56 \mathrm{kWh} / \mathrm{m}^{2} /$ day) were applied as the solar input. Results show that the hybrid configuration eliminates diesel use for low and medium inflows, while under extreme inflow, annual diesel consumption is reduced to $\approx \mathbf{8, 3 0 0 ~} \mathrm{L}$ compared with $\approx 8,950 \mathrm{~L}$ for a diesel-only baseline, yielding $\approx 7 \%$ fuel savings. The system achieves renewable energy fractions of $\approx 52 \%$ (low), $17 \%$ (medium), and $7 \%$ (extreme) and maintains monthly photovoltaic contributions of 60$75 \%$ even during variable rainfall. Beyond system modeling, the study incorporates a data-engineering pipeline that enables sensor data acquisition, storage, and visualization, with potential integration of artificial intelligence (AI) for predictive scheduling and anomaly detection. This framework demonstrates the role of applied AI and data engineering in advancing sustainable water-energy management in mining, aligning with Industry 5.0 principles and the broader goals of sustainable development.

Abstract. Prastyaningsih SR, Hardiwinoto S, Musyafa, Prehaten D, Effendy I, Ridwan N, Aini FK, Benatar GV. 2025. Termite diversity and infestation levels on Eucalyptus pellita plantations in Riau, Indonesia. Biodiversitas 26: 5495-5504. Eucalyptus pellita is used in industrial plantations due to its high productivity. However, monoculture plantations are vulnerable to termite infestations, which can reduce wood volume. This study aimed to assess termite infestation in E. pellita plantations in Riau, Indonesia. A total of 5,791 trees were surveyed across 24 plots representing different stand ages to determine the incidence and intensity of termite damage. Specimens from infested trees were identified morphologically and molecularly using the mitochondrial cytochrome oxidase subunit II COII gene marker and phylogenetic analysis was applied to confirm species-level identification. Three termite species were identified: Coptotermes sepangensis, Nasutitermes matangensis, and Microcerotermes havilandi. Infestation was recorded in 3-4-year-old stands, with 5th rotation plots exhibiting greater severity than 4th rotation plots; among all species, Microcerotermes caused the greatest damage, followed by Nasutitermes and Coptotermes. Across all stands, 301 trees (5.2%) were found to be infested. Damage severity was assessed by observing trunk conditions, especially when termite mounds covered half or the entire diameter, and volume loss was estimated using Huber’s formula. The highest infestation intensity (21.5%) occurred in the 4th rotation, although most cases were classified as mild (scale 1). In contrast, the 5th rotation showed higher infestation severity, with 12.3% of trees severely damaged (scale 4), resulting in the greatest wood volume loss. In 5th rotation, three-year-old tree showed the highest loss of volume, estimated at 9.10%, or 7.54 m³/ha. This is the first molecular confirmation (COII) report of termite infestation on E. pellita in Riau. The results provide baseline information on species composition, infestation patterns and associated wood loss in E. pellita plantations.

Stunting remains a significant public health issue in Indonesia, particularly in Aceh Province, where the prevalence reached 33.18% in 2021 exceeding the WHO threshold. In North Aceh District, 14.3% of children under five were stunted, with Simpang Keramat District reporting one of the highest rates (35%). This study aimed to evaluate the effectiveness of a combined intervention Moringa leaf-supplemented food within a balanced diet, baby massage, and Tui Na massage in improving the nutritional status of stunted children aged 0–2 years. A quasi-experimental pre-post design was used, involving 50 stunted children divided equally into treatment and control groups. The intervention was conducted over 2.5 months. Paired t-test results showed significant increases in weight (t24 = -6.656, p < .000) and height (t24 = -3.244, p < .003) in the treatment group compared to the control group. These findings suggest that combining dietary fortification using locally available Moringa leaves with physical stimulation through massage therapy can effectively enhance growth outcomes. Given the multifactorial nature of stunting, this study recommends that nurses and health practitioners incorporate both nutrition-based interventions and sensory-stimulating therapies to address growth delays in young children, particularly in high-risk areas such as Aceh.

This article report Cd(1-x)SO4:XDy3+ phosphor which were synthesized using recrystalization method, Crystal structure and phase formation of the product was identified by powder XRD technique and partical morphology was studied by FE-SEM .According to the results of the PL analysis carried out by the Hitachi F-7000 fluorescence spectrophotometer. The luminescence decay curves of CdSO4(1-x):xDy3+ phosphors for utilizing excitation wavelengths of 350nm and emission wavelengths of 575 nm that were recorded by and HITACHI F-7000 Spectroflurophotometer. It has been shown that the decay time is around 0.75 ms The photoluminescence (PL) spectra were investigated under 350 nm (UV) excitation, two emission peaks were obtained one at 483 nm, which falls under the blue region, and the other at 575 nm, which falls under the bright yellow region. These two emission peaks correspond to the 4F9/2 to 6H15/2 and 4F9/2 to 6H13/2 transitions, respectively. We suggest that the CdSO4:Dy3+ can be employed as an excellent luminescent material with ultraviolet radiation as an excitation source for a variety of optical applications based on its optical properties. White light for WLED applications can be produced by a special phase doping of Dy3+ species in the host material with yellow and blue emission.

Background: Diabetes mellitus (DM) is a metabolic disorder characterized by elevated blood glucose levels, resulting from the body’s inability to produce sufficient insulin or to utilize insulin effectively. This condition leads to hyperglycemia, which manifests through typical symptoms such as frequent urination in large volumes with a sweet odor, commonly referred to as “sweet urine.” A blood glucose concentration exceeding 200 mg/dL indicates an abnormality in the body’s glucose regulation. Objective: This study aims to identify the risk factors contributing to the incidence of diabetes mellitus in coastal areas. Methods: This research employed a systematic literature review (SLR) design, utilizing a structured library research approach. References were collected from secondary data derived from previous studies. The data sources consisted of scientific articles and journals accessible through the Google Scholar database. Results: The review revealed that factors such as age, sex, occupation, dietary patterns, physical activity, genetic predisposition, and lifestyle were significantly associated with the incidence of DM in coastal areas. In contrast, dietary habits related to specific food consumption and smoking behavior did not show a significant association with DM incidence in these regions. Conclusion: Risk factors contributing to diabetes mellitus in coastal areas include individual characteristics, family history, lifestyle, fruit consumption, and physical activity levels. Conversely, vegetable consumption and general dietary patterns were not identified as primary risk factors. Keywords: diabetes mellitus, characteristics, genetics, physical activity, lifestyle, coastal areas

Kehamilan risiko adalah kehamilan yang akan menyebabkan terjadinya bahaya dan komplikasi yang lebih besar baik pada ibu maupun pada janin dalam kandungan serta menyebabkan kematian, kesakitan,kecacatan,dan ketidak nyamanan. Kehamilan risiko tinggi ditemukan pada ibu hamil yang terlalu tua (diatas 35 tahun), terlalu muda (di bawah 20 tahun), terlalu banyak (lebih dari 4 kali), dan terlalu dekat (jarak melahirkan kurang dari 2 tahun) atau lebih dikenal dengan 4 terlalu (4T). Tujuan penelitian untuk mengetahui hubungan penggunaan kontrasepsii dengan kejadian kehamilan berisiko diwilayah puskesmas mattirobulu kabupaten pinrang tahun 2024. Metode penelitian ini merupakan jenis penelitian kuantitatif dengan menggunakan desain cross sectional, jumlah sampel 35 responden, mengumpulkan data lewat kusioner dan observasi. Hasil penelitian berdasarkan hasil uji statistic di peroleh nilai P-Value=0,001 maka Ho di tolak artinya ada hubungan antara penggunaan kontrasepsi terhadap kejadaian kehamilan berisiko empat terlalu banyak . Kesimpulan diharapkan bagi petugas kesehatan dipuskesmas untuk memberikan upaya edukasi dan pomotif pada masyarakat menggenai kejadian kehamilan berisiko beserta upaya pencegahannya.