Center for Scientific Publication
Not a member yet
4029 research outputs found
Sort by
Perbandingan Parameter Hidraulik Perencanaan Bendung Gerak Kantelstuw Dengan Bendung Tetap Di Bendung Doorbraak, Almelo, Belanda
Full text linkAbstract— The Doorbraak Weir is located in the De Doorbraak Diversion, channeling water from the Loolee River to the Eksosche Aa River. This weir is a fixed cascade type with a height of 2.1 m, serving flood control as well as livestock and agricultural needs. The required water level is +10.5 NAP, but this is often not met during the summer because the weir crest elevation is only +10.3 NAP. To address this, a plan is proposed to convert the weir into a movable kantelstuw or tilting gate. The proposed gate has a crest elevation of +10.5 NAP, with dimensions of 0.5 m x 1.5 m and a gate base height of 0.8 m. Analysis shows that the movable weir is safer, with a Froude number of 2.5 and a jump length of 1.92 m, compared to the fixed weir. The gate operation scenario meets the minimum freeboard requirement of 0.90 m, and the levee is proven to be safe from backflow. Based on hydraulic parameter comparisons, the movable kantelstuw type weir is recommended for the Doorbraak Weir construction
Comparison of Ensemble Learning Methods in Classifying Unbalanced Data on the Bank Marketing Dataset
Full text linkThe banking industry is experiencing rapid growth, particularly in telemarketing strategies to increase product and service sales. Despite widespread use, these strategies need higher success rates due to data imbalance, where fewer customers accept offers than those who reject them. This study evaluates machine learning algorithms, including Random Forest, Gradient Boosting, Extra Trees, and AdaBoost, without and handling imbalanced data using the Random Over-Sampling Examples (ROSE) method. The evaluation covers accuracy, precision, recall, F1-score, and AUC of the ROC curve. Results indicate that Random Forest and AdaBoost consistently perform well, with Random Forest maintaining a high accuracy of 91.00% after handling imbalanced data. Gradient Boosting and Extra Trees improve in precision post-oversampling. All models exhibit high AUC values, close to 0.94, demonstrating excellent differentiation between positive and negative classes. The study concludes that addressing data imbalance enhances model performance, making these models suitable for effective telemarketing strategies in the banking sector
Prediction of Nike’s Stock Price Based on the Best Time Series Modeling
Full text linkNike is one of the world's largest shoe, clothing, and sports equipment companies. The more modern the development of the era, the more diverse the fashion. Of course, investors can consider this when deciding whether to invest in Nike's brand shares. Stock prices constantly fluctuate up and down, so investors need to implement strategies to minimize losses in investing to achieve economic growth. This supports the Sustainable Development Goals (SDGs) in point 8 regarding the importance of sustainable economic growth and investment in infrastructure development to improve economic welfare. Investors can minimize losses by predicting or forecasting stock prices. Stock prices can be analyzed using specific methods. The update that will be brought in this study is the Nike brand stock price prediction for the 2020-2024 period using the best model from the time series method comparison conducted using classical nonparametric, which consists of the kernel estimator method and the Fourier series estimator method and modern nonparametric using the Support Vector Regression (SVR) method. Based on the analysis method, the best method is selected through the minimum MAPE value. A comparison of the results of Nike brand stock price predictions using several methods shows that the MAPE value of the Nike brand stock price data analysis is the minimum obtained using the kernel estimator approach, which is 1.564%. Thus, the kernel estimator approach predicts the Nike brand stock price much better. Predictions using the best methods can be recommendations and evaluations for economic actors to prepare better economic planning
Alternative Planning Slope Stabilization on Roadway Case Study : Landslide on Tabone-Polewali Road Section KM 168+790 to 168+820
No full textThe Tabone-Polewali road section is a national road in the West Sulawesi Province that connects the Polewali Mandar Regency and Mamasa Regency. The terrain along this section consists of fairly steep slopes. In 2021, landslides occurred at several points along the Tabone-Polewali road, one of which happened at KM 168+790 to 168+820 during heavy rain. The cause of the landslides is estimated to be water flowing from the upper slope to the lower slope without effective water runoff management. No measures have been taken at this location, making it still hazardous for road users. There are several alternative designs for permanent mitigation, including bored. Slope stability existing analysis was conducted using the Limit Equilibrium Method and Finite Element Method (FEM) for slope reinforcement stability analysis. The results obtained from this study show that the Safety Factor for the existing condition is 1,09. The Safety Factor for the alternative design using bored piles is 1,277 where this has met the minimum safety factor value requirements
CONCEPT OF ORNAMENTATION BASED ON LOCAL WISDOM IN PUBLIC BUILDINGS IN THE URBAN AREA OF TAKENGON
Full text linkThe Gayonese, an ethnic group indigenous to the highland regions of Aceh, are distinguished by their rich cultural heritage. Their cultural patrimony is characterized by a diversity of artistic expressions, including ornamental designs that embody the community's local wisdom and cultural identity. Takengon, the most vibrant town in Aceh's highland region, is home to numerous public buildings, including governmental, religious, educational, hospitality, and other service facilities, which incorporate ornamental elements as expressions of local wisdom. However, the application of these ornaments varies significantly by placement, symbolism, and color due to the absence of standardized guidelines for their implementation in public architecture. The main objectives of this research are twofold. Firstly, to examine the prevailing patterns of ornamentation in public buildings from a semiotic perspective. Secondly, grouping in local wisdom to develop a comprehensive framework for the systematic integration of ornamentation within Takengon's architectural landscape. This research employs a qualitative approach with semiotic analysis to interpret the visual symbols and meanings of ornamentation in public buildings, aiming to reveal the cultural and symbolic values embedded within. The methods of data collection employed involve the observation of ten public buildings, documenting, and interviewing informants. The informants in this research were government officials, cultural or community leaders, building consultants, and academics. The findings reveal the placement of ornamentation typically occurs in three main zones: the head, body, and base of the building. In terms of ornament typology, natural motifs are predominant, surpassing geometric and floral patterns, with no representations of fauna identified among the ten sampled buildings. Ornamentation in Takengon's public buildings primarily serves a decorative purpose, classified as applied ornamentation, rather than fulfilling structural or constructive functions. The contribution of this research lies in its enrichment of the discourse on cultural identity in the built environment, particularly in relation to ornamentation
Estimation Earthquake Source Parameters in Mentawai Island Region Using Moment Tensor
Full text linkMentawai Islands is a seismically active zone that often experiences earthquakes due to the interaction of tectonic plates, so it is necessary to analyze the earthquake source parameters to understand its characteristics. This research aims to determine the earthquake source parameters in the Mentawai Islands in 2023 for earthquakes with magnitude Mw ≥ 5.5. Three-component waveform data from GEOFON network is used for moment tensor inversion process in time domain. Waveform data used in the frequency range of 0.01 to 0.025 Hz with a match value of calculation and observation data worth >70%. The results of waveform data analysis show that the earthquake in the study area was caused by tectonic activity characterized by the percentage of Double-Couple (DC) more dominant than Compensated Linear Vector Dipole (CLVD), with shallow depth. In addition, the focal mechanism of the waveform data analysis indicates that the type of fault that causes the earthquake in the study area is a reverse fault. This result is consistent with earthquake catalogs (GCMT, USGS, and GFZ)
Implementasi Corporate Social Responsibility dalam Mendukung Kemandirian Masyarakat Lokal
Full text linkThis study analyzes the implementation of Corporate Social Responsibility (CSR) by SIG Pabrik Tuban in supporting local community independence. CSR is not only a corporate social responsibility but also a strategy to build harmonious relationships with the community and ensure economic sustainability. A descriptive qualitative method is applied in this research, with data collected via interviews, observations, and literature studies. The findings indicate that the company's CSR programs focus on economic empowerment, capacity building, and infrastructure development with a strategic allocation of funds. Through collaboration with the Forum Masyarakat Kokoh (FMK), CSR programs are designed based on community needs to enhance effectiveness and sustainability. CSR implementation has been proven to improve community skills, expand economic access, and strengthen the local business ecosystem. With a participatory and systematic approach, CSR at SIG Pabrik Tuban plays a crucial role in fostering a more independent and empowered society. These efforts highlight the importance of corporate involvement in sustainable development and community welfare
Mathematical Modelling of Longitudinal Vibration on Propulsion System 5200 DWT General Cargo Ship
Full text linkThe vibration level of the propulsion system will change during its operation. This vibration is caused by harmonic excitation forces produced by the rotation of the main engine and propeller shaft. Ship propulsion systems experience longitudinal, torsional and lateral vibrations. Excessive vibration will produce noise and reduce engine performance. Vibrations can also cause resonance in the system, which can be fatal and damage the structure. The excitation frequency value is close to or equal to the natural system frequency, which causes resonance. This paper has identified the vibration response of the propulsion system by using numerical software through mathematical modelling governed by ABS. In addition, the total vibration response was obtained using the modal analysis method by summing up the contributions of each mode. The excitation source generated is due to the rotation of the main engine. Ultimately, the response obtained will be adjusted to the standard class. The modelling results obtained a 3-Degree-of-Freedom forced vibration model consisting of three masses and three springs. The resulting response values are displacement and velocity, where the highest response occurs at 347 rpm with a deviation of ±0.1345 mm to ±0.3371 mm and a velocity value of ±4.8847 mm/s to ±12.2424 mm/s. The slightest response occurs at 459 rpm with a deviation range of ±0.0034 mm to ±0.0050 mm and velocity values of ±0.1634 mm/s to ±0.2382 mm/s. Based on all the results of adjusting the vibration response value with the ABS class vibration limit graph, the vibration is still below the permissible threshold line
Estimation of Stunting and Wasting in Sumatra 2022 with Nadaraya-Watson Kernel and Penalized Spline
Full text linkThis study aims to estimate the prevalence of Stunting and Wasting in Sumatra in 2022 using nonparametric regression methods, specifically the Nadaraya-Watson Kernel and Penalized Spline regression models. Both models were applied to assess the relationship between these two correlated response variables and various predictor variables, such as low birth weight, sanitary facilities, poor population, and exclusive breastfeeding. The results showed that the Nadaraya-Watson Kernel regression, particularly using the Gaussian kernel, provided the best fit with minimal prediction error, as indicated by its low Generalized Cross-Validation (GCV) value of 0.024 and high R-squared values (0.9992 for Stunting and 0.9995 for Wasting). In contrast, the Epanechnikov kernel and Biweight kernel produced higher GCV values (0.110 and 0.356, respectively), indicating less optimal performance. For the Penalized Spline model, optimal parameters were determined with a smoothing parameter λ of 5 and 3 knots, which balanced model flexibility and smoothness. This research underscores the potential of nonparametric regression techniques in capturing complex relationships in health data and provides insights for improving interventions aimed at addressing child malnutrition in Indonesia
Application of Electrocoagulation in Stabilizing pH and Removing Pollutants from Domestic and Urban Wastewater Using Aluminum Electrodes
Full text linkThis study investigates the application of electrocoagulation using aluminum electrodes to stabilize pH and remove pollutants from various real wastewater sources in Surabaya, Indonesia, including laundromats, fast food restaurants, campus canteens, mechanical workshops, and tofu industries. A batch-mode electrocoagulation process was conducted using 10 A of current over 60 minutes, and the performance was evaluated based on pH changes, turbidity, and Total Suspended Solids (TSS) removal. The highest TSS removal efficiency of 98% was observed in machine shop wastewater, while laundromat samples also achieved high performance with 97% reduction. Electrocoagulation demonstrated strong pH stability in effluents with buffering compounds such as those from campus canteens and workshops, whereas fluctuating pH and turbidity levels were found in tofu and fast-food effluents, indicating the need for process optimization. Turbidity decreased significantly across all samples, confirming effective floc formation during electrolysis. The findings confirm that electrocoagulation is a promising method for treating both domestic and urban wastewater. However, integrating additional processes such as adsorption or filtration is recommended to enhance performance for complex wastewater compositions and ensure compliance with discharge standards