International Journal of Engineering, Technology and Natural Sciences

Assessment of Fire Safety Preparedness in Elementary Schools: Evidence from Banjarmasin, Indonesia

Madschen Sia Mei Ol Siska Selvija Tambun — 2026-07-01

School fire incidents present critical risks to vulnerable demographics, particularly in developing urban areas. This study evaluates the structural fire preparedness of primary education facilities. Employing a descriptive qualitative research design, this study selected thirty-two elementary schools in Banjarmasin City through purposive sampling, prioritizing institutions near high-density residential zones with predominantly wooden structures. The assessment framework was established based on the Banjarmasin City Regional Regulation Number 13 of 2008, supplemented by the Indonesian Ministry of Public Works Regulation Number 26/PRT/M/2008. The evaluation criteria strictly measured four core dimensions of fire preparedness: active fire protection facilities, passive fire truck access, life-saving evacuation facilities, and Building Fire Safety Management (MKKG). Primary data were gathered through direct on-site observations, standardized checklists, and interviews with school principals. The findings revealed severe gaps across the evaluation criteria: only 22% of schools possessed functional fire extinguishers, none had fire hydrants, and life-saving assets such as emergency lights and exit signs were completely absent. While 94% of the sites demonstrated sufficient fire truck accessibility, the implementation of formal MKKG scored 0% across all sampled institutions due to the total absence of emergency response organizations or written evacuation plans. It is concluded that fire safety readiness remains critically suboptimal. These systematic deficiencies highlight an urgent need for policy interventions, integrating school operational budgets with standardized emergency response governance.

Fake News Detection on News Websites Using Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT)

Niar Yuniarti — 2026-07-27

The rapid spread of misinformation in digital media has become a serious challenge, particularly in the context of Indonesian online news consumption. Therefore, this study aims to develop an effective fake news detection system by leveraging a contextual language model tailored to the Indonesian language. This study proposes the use of the Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT) model to detect fake news by optimizing the model architecture and hyperparameter settings. IndoBERT, as a contextual language model, functions to capture the nuances of language in text and can be adjusted at the output layer for classification tasks. In this study, the model architecture is enhanced by adding classification layers, including a linear layer, ReLU activation function, dropout, and a sigmoid output layer, which is expected to improve the accuracy of fake news detection. This study uses Indonesian news data from two sites, Turnbackhoax and Detik.com, covering various topics. The model is trained with different hyperparameter settings, including batch size, dropout, and learning rate, to find the best combination that achieves high accuracy in detecting fake news. The trained model is integrated into a Flask-based web application, allowing users to automatically verify the truth of news through a simple and user-friendly interface. The results show that the best combination, with a learning rate of 2e-5, dropout of 0.1, and a batch size of 32, achieves an accuracy of 96.4%. This study demonstrates that IndoBERT, with the appropriate settings, can be effectively applied to detect fake news with high accuracy, contributing to the development of hoax detection applications.

Selective Catalytic Conversion of Lignocellulosic Biomass into Furan Derivatives Using Barium Chloride

Abubakar M. Ali — 2026-07-29

The growing demand for sustainable chemical production has intensified interest in biomass-derived platform chemicals. This study investigates the thermal methanolysis of Gmelina arborea leaves, an abundant non-edible lignocellulosic biomass, for the selective production of furfural and 5-methylfurfural (5-MF). Using barium chloride (BaCl₂) as a Lewis acid catalyst in methanol medium, the process was conducted under mild conditions (60 ^oC, atmospheric pressure) across reaction times of 10 - 60 min. GC-MS analysis revealed that furfural yield peaked at 3.91% (143.7 mg/g) after 30 min, while 5-MF reached 2.78% (102.5 mg/g) at 20 min. The distinct temporal profiles highlight the influence of reaction kinetics and thermal sensitivity on product selectivity. Statistical analysis confirmed reaction time significantly affected yields (p < 0.05), and reproducibility assessment showed excellent precision (RSD < 2.13%). Mechanistic insights suggest Ba²⁺ ions facilitate selective glycosidic bond cleavage in hemicellulose, promoting sugar dehydration to furan derivatives, with methanol suppressing polymerisation side reactions. This work offers a low-energy, acid-moderated route for biomass valorisation, addressing key limitations of conventional methods, corrosive acids, high temperatures (>150 ^oC), and pressurised systems. The results demonstrate that BaCl₂ effectively catalyses solvolytic hydrolysis and dehydration with high selectivity while minimising degradation pathways. This study underscores the potential of Gmelina arborea as a renewable feedstock in green chemical manufacturing and contributes to sustainable biorefinery development by demonstrating an environmentally benign catalytic system for converting underutilised plant residues into valuable platform chemicals, supporting the transition toward a circular bioeconomy.

Urbanization and Microclimate Change: A Case Study of Bukit Halimun, Luwuk

Muhammad Faqih Imawan Hipan — 2026-07-29

Urbanization is a process that occurs when the population in urban areas continues to rise, often accompanied by the development of infrastructure and other constructions within the urban scope. While urbanization brings various benefits such as economic growth and technological advancement, it also has negative impacts on the urban microclimate. These include the continuous reduction of green spaces leading to higher temperatures in the area and creating discomfort for humans, air pollution, declining air quality in urban environments, and potential health risks. Cities are expected to accommodate various activities of their residents, including housing, work, and recreation. With limited land available and a growing population, the demand for land increases. This situation inevitably leads to changes in land use within urban areas, where green spaces shrink over time. This study employs a deductive quantitative method with an evaluative approach, utilizing object modelling, simulation, and analysis through the Envi-MET software. This method is used to assess the impact of urbanization on microclimate changes, considering variables such as air temperature, wind speed, and relative humidity, which can affect outdoor thermal comfort levels. The research is focused on the Bukit Halimun Area, Luwuk City, Banggai Regency, Central Sulawesi. In 2007, the local government of Banggai Regency issued a policy for the Building and Environmental Planning for the Bukit Halimun area, covering a total planning area of 92 hectares. This area is intended for the development of office, residential, commercial, and educational zones.

Applying the HIRADC Method to Occupational Safety Risk Assessment and Control in a Private Higher Education Institution

Soerahman Soerahman — 2026-07-30

Occupational health and safety in higher education institutions requires systematic management because academic, laboratory, administrative, and mobility-related activities may generate physical, chemical, biological, and ergonomic hazards. This study aims to identify hazards, assess risk levels, and formulate priority safety controls in a campus environment using the Hazard Identification, Risk Assessment, and Determining Control (HIRADC) method. The research was conducted at a private higher education institution in Tangerang using a mixed-method approach involving field observation, interviews, and questionnaires administered to 50 respondents. Risk assessment was performed by determining likelihood and severity scores, which were then classified into low, moderate, high, and extreme categories. The findings identified 15 potential hazards, 5 high risks, 5 moderate risks, and 5 low risks. Stair areas, pharmaceutical laboratories, and engineering laboratories were identified as the dominant risk locations. Recommended controls include improving lighting, installing safety signs and posters, providing fire extinguishers, ensuring the use of personal protective equipment, conducting safety education, and implementing periodic evaluations. The findings contribute to the development of a systematic occupational safety management framework for higher education institutions and provide practical guidance for prioritizing risk control in campus environments.

Design, Considerations, and Implementation of a Solar-Powered Smart Street Lighting System with Integrated CCTV and IoT-Based Monitoring

Muhammad Rizani Rusli — 2026-07-30

This study presents the design, implementation, and short-term experimental validation of a standalone solar-powered smart street lighting system integrating LED illumination, CCTV surveillance, and Internet of Things (IoT)-based real-time monitoring within a single pole-mounted platform. Load profiling identified a total energy demand of 1176 Wh/day, which increased to 1529 Wh/day after applying a 30% design margin. These requirements guided the selection of a 460 Wp monocrystalline photovoltaic module, a 40 A maximum power point tracking charge controller, and a 12.8 V, 150 Ah LiFePO₄ battery. Laboratory testing demonstrated the expected bulk/MPPT, boost, and float charging stages, while accelerated battery testing at 0.27C measured a capacity of 151 Ah. During five consecutive days of field testing under sunny to partly cloudy conditions, the post-charging state of charge (SOC) ranged from 90.0% to 97.2%, while the next-morning SOC ranged from 58.6% to 68.4%. The average nighttime SOC reduction was 32.8%, with a range of 25.1% to 37.3%. The IoT platform provided real-time monitoring of photovoltaic generation, battery condition, and load operation, while maintaining communication availability above 99% during the observation period. These results demonstrate short-term autonomous operation and technical feasibility under the observed field conditions. Longer seasonal testing and a complete life-cycle economic assessment remain necessary to evaluate long-term reliability and financial feasibility.

Ensemble Deep Learning with Attention Mechanism and Explainable AI for Enhanced Brain Tumor Classification from MRI Images

Krisna Nuresa Qodri — 2026-07-30

Brain tumors represent a significant clinical challenge, with accurate classification being essential for treatment planning. Current deep learning approaches face two critical limitations: insufficient robust-ness across diverse imaging conditions and lack of clinical interpretability. Here we present an ensemble deep learning framework integrating three complementary architectures (MobileNetV2, EfficientNetB3, and DenseNet121) with spatial attention mechanisms and explainability features. Using 7,023 MRI images across four diagnostic categories (glioma, meningioma, pituitary tumor, and tumor-absent), our approach achieved 98.47% classification accuracy with balanced performance across all classes (F1-scores: 98.12% for glioma, 97.89% for meningioma, 98.76% for pituitary, 99.21% for tumor-absent cases). The ensemble demonstrated statistically significant improvement over individual models (p < 0.01, McNemar’s test), with gains of 1.24-1.75 percentage points. Integration of Gradient-weighted Class Activation Map-ping provided interpretable visual explanations with activation patterns consistently focusing on tumor regions. The findings demonstrate the potential of combining ensemble learning, attention, and visual explanation for brain tumor classification. However, the results represent internal validation on a single partition of aggregated public datasets and require confirmation through repeated validation and independent clinical evaluation.

Optimization of Road Worthiness Assessment Using a Star Rating Scale

Louise Elizabeth Radjawane — 2026-07-30

The Perintis Kemerdekaan Road serves as a primary arterial road and is a national road, providing access to various public facilities such as offices, schools, places of worship, general hospitals, and shopping centers. A 6/2T-type road located in an area with mixed land use has a high level of service complexity because it must accommodate large traffic flows as well as roadside activities from various land uses. In such conditions, an evaluation of the road’s worthiness assessment for purpose is necessary to ensure that the roadway still meets technical and administrative requirements and remains safe for public use. This research objective is to determine the star rating for safe road infrastructure on a 6 lanes 2 lines divided road type and the relationship between the star rating value and the risk factor score for accident types on the road. The research method is the Technical Guidelines for Road Worthiness 11 PBM 2023 as a reference. The assessment of the star rating shows that the Perintis Kemerdekaan Road at KM 12, length 1 kilometer and divided into 100-meter segments, has safer road categories in segments 1, 4 to 10 (Star rating scale rating 4), and segment 2 (Star rating scale rating 5), while segment 3 is included in the safe road category. On urban roads with mixed land uses, the highest risk of property access accidents occurs in lanes less than 3.25 meters wide, with inadequate pedestrian facilities, and the absence of school safe zones. Run-off (driver-side) accidents occur when lane width is insufficient, road surface roughness is inadequate, and delineation are inadequate. For future research, integrating social aspects such as accessibility and economic aspects such as maintenance costs and local economic benefits in assessing the functional worthiness of roads.

From Tendering to Governance: Exploring e-GP Implementation Barriers

Md. Abdur Rashid — 2026-07-29

Electronic Government Procurement (e-GP) has become a critical digital governance mechanism for enhancing transparency, accountability, efficiency, and competitiveness in public procurement. However, the effectiveness of e-procurement depends not only on technological implementation but also on institutional readiness, operational capacity, and supportive regulatory frameworks. This study investigates the implementation challenges of the e-GP system within the Roads and Highways Department (RHD) of Bangladesh and identifies strategies for improving procurement governance. A qualitative research approach was employed through Key Informant Interviews (KIIs) involving 206 Procurement Entity (PE) officers representing various procurement roles, including PE users, Tender Opening Committee (TOC) members, administrators, and Tender Evaluation Committee (TEC) members. The findings reveal that although e-GP has improved procurement digitalization, approximately 86% of respondents continue to experience challenges related to procurement time, operational processes, system functionality, and institutional procedures. Key challenges include lengthy tender evaluation processes, excessive documentation requirements, and delayed financial clearance, incomplete digitalization of evaluation activities, weak document verification, and coordination difficulties among procurement committees. The study also identifies governance challenges, including limited participation of small and medium contractors, and continued dependence on physical documentation. These findings indicate that digital procurement transformation requires more than technological adoption; it requires integrated institutional and governance reforms. The study recommends strengthening system reliability, automating procurement processes, improving effective banking integration, expanding user training, and developing competitive procurement mechanisms to transform e-GP into a comprehensive digital procurement governance system.

Rebuilding Mobile Cold Storage System to Maintain Fishery Product Quality

Mustopa Kamal — 2026-08-27

The inadequate performance of the refrigeration compressor can limit the ability of a mobile cold storage system to maintain stable low temperatures during continuous operation, potentially affecting its application for fishery-product storage and distribution. This study aims at evaluating the technical performance of a rebuilt mobile cold storage system following the replacement of its original automotive air-conditioning compressor with a 1 HP air-conditioning compressor. An experimental case-study method was conducted using a 1,560-L mobile cold storage unit located at the Center for Testing the Application of Marine and Fisheries Products in East Jakarta. The rebuilding procedure comprised initial system inspection, removal of the existing compressor, modification of the compressor mounting, installation of the 1 HP compressor, reconnection of the suction and discharge refrigerant lines, electrical-system adjustment, system evacuation, refrigerant charging, leak inspection, and operational testing. Cooling performance was evaluated from the temperature reduction during an 8-h experimental test using 10 L of water as the test medium, with an initial temperature of approximately 30°C. The rebuilt system reduced the test-medium temperature from approximately 30°C to -4°C within 8 h and maintained stable cooling operation throughout the test period. These results demonstrate that replacement of the automotive air-conditioning compressor with a 1 HP air-conditioning compressor restored the cooling capability of the 1,560-L mobile cold storage system. The findings indicate that targeted compressor replacement, supported by appropriate mechanical, refrigerant-piping, and electrical modifications, provides a practical component-level rebuilding approach for restoring existing mobile cold storage systems used in fishery-product applications.