cover
Contact Name
Christian Harito
Contact Email
christian.harito@binus.edu
Phone
+6221-5350660
Journal Mail Official
aagung@binus.edu
Editorial Address
Universitas Bina Nusantara Jl. Kebon Jeruk Raya No.27 Kebon Jeruk, Jakarta Barat 11530
Location
Kota adm. jakarta barat,
Dki jakarta
INDONESIA
Engineering, Mathematics and Computer Science Journal (EMACS)
ISSN : -     EISSN : 26862573     DOI : https://doi.org/10.21512/emacs
Engineering, MAthematics and Computer Science (EMACS) Journal invites academicians and professionals to write their ideas, concepts, new theories, or science development in the field of Information Systems, Architecture, Civil Engineering, Computer Engineering, Industrial Engineering, Food Technology, Computer Science, Mathematics, and Statistics through this scientific journal.
Articles 199 Documents
Avoided Emissions and Rebound Robustness in Data Center Heat Reuse Andrae, Anders S.G.
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16055

Abstract

Data centers are becoming increasingly important components of modern digital infrastructure, but their growing electricity demand has raised concerns regarding their environmental footprint. At the same time, data centers generate substantial quantities of low-grade waste heat that may be reused in nearby buildings or district heating systems. Such applications have attracted growing interest as examples of ICT-enabled heat-substitution solutions. However, methodologies for quantifying the associated avoided emissions remain under development, particularly in relation to emerging standards and guidance documents. This study applies internationally-aligned avoided-emissions frameworks to a published case study of direct waste heat reuse from a free-air-cooled data center. The engineering basis is derived from a previously published simulation of a 1 MW data center operating without heat pumps and reusing excess heat directly. The avoided-emissions calculation follows a standardized formulation including emissions from conventional heat production and emissions associated with the waste heat reuse solution.  For the case with an Energy Reuse Factor (ERF) of 0.59 and a 40°C exhaust temperature, the avoided emissions are estimated to around 783 tCO₂e per year. Moreover, a new rebound robustness metric is proposed to quantify the magnitude of rebound effects required to overturn the avoided-emissions conclusion. The results indicate a break-even rebound of nearly 100% of the useful heat service, suggesting that the avoided-emissions claim is highly robust. The study reveals how ICT-enabled heat reuse can be assessed consistently using internationally standardized concepts and guidance while introducing a complementary break-even rebound and rebound robustness indicator for future research.
Comprehensive Analysis of TF-IDF and Word2Vec-Based Text Feature Representation on Support Vector Machine Performance in Sentiment Analysis of Telemedicine Reviews Lestari, Citra Ayu Tri; Khudori, Ahsanun Naseh; Pradini, Risqy Siwi
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16081

Abstract

User reviews of telemedicine applications provide valuable insights into user satisfaction and service quality. Accurate sentiment analysis of these reviews depends greatly on the text representation method used before classification. Although TF-IDF and Word2Vec have been widely applied in sentiment analysis, their comparative performance on Indonesian-language telemedicine reviews with a highly imbalanced sentiment distribution has not been sufficiently investigated. So, in order to analyze the sentiment of Halodoc application evaluations using the Support Vector Machine (SVM) classifier, this study analyzes the performance of TF-IDF and Word2Vec, two text representation approaches. The Google Play Store was combined from January 2024 to October 2025 for a grand total of 10,443 reviews. After implementing user-rated automatic sentiment labeling, 6,574 valid reviews were obtained using Indonesian-specific preprocessing that included case folding, cleaning, stopword removal, and stemming. Stratified 5-fold cross-validation using recall, accuracy, precision, and F1-score metrics was used to test the models. Word2Vec only managed 75% accuracy, 78% precision, 79% recall, and 78% F1-score in the experiments, but TF-IDF achieved 90% accuracy, 89% precision, 90% recall, and 89% F1-score. TF-IDF outperformed Word2Vec because frequency-based features are more effective in representing short review texts, whereas Word2Vec generally requires a substantially larger corpus to learn robust semantic representations. These findings provide empirical evidence for selecting appropriate text representation methods in Indonesian telemedicine sentiment analysis and contribute practical guidance for developing more accurate digital healthcare opinion mining systems.
The Impact of Ensemble Stacking with Machine Learning and Statistical Models on Improving Sales Forecasting Accuracy Yohana, Prima; Gunawan, Alexander Agung Santoso; Ngarianto, Heri
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16162

Abstract

Sales forecasting in retail remains a challenge, as changing customer behavior and seasonal patterns undermine predictive accuracy. The Ensemble Stacking Framework for Machine Learning & Statistical Models is being suggested to improve forecasting of dynamic demands. For example, during promotional events, managers adjust their plans to account for changes in weather or unexpected changes in consumer preferences. They are unable to predict these changes. An Ensemble Stacking Framework was created using a Meta-Learner (RidgeCV) trained on historical retail sales data and several base learners (Random Forest, Extra Trees, XGBoost, and Multiple Linear Regression). Tree-Based Models help to create non-linear relationships, while Linear Regression helps us understand trend information. The Meta-Learner learns how to weigh each model based on the error each model makes over time. We assessed the quality of our model’s performance through RMSE and MAE. Special attention was given to assessing how well our models performed during periods of high volatility (i.e. holidays and post-promotional dips). Our model performed better than all other Base Models at all times. Also, stacked predictions showed less variation week-to-week and were able to return to normal quicker than other models after sudden shifts in demand. In practice, it allows for more confident safety-stock decisions and timely replenishment adjustments. Retailers can run a number of models in parallel and continuously rebalance them with new data as it comes in. Future work may include online updates of meta-learners and external covariates such as local events or competitor pricing.
Walk-Forward Evaluation of ROI-Based Decision Thresholds for Random Forest Loan Default Prediction Lucky, Henry; Anderies, Anderies
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16455

Abstract

Default prediction models for peer-to-peer lending are often tested on random data splits and judged by accuracy, which can hide temporal leakage and says little about investor returns. This study evaluates a random forest on 1,020,768 resolved 36-month Lending Club loans issued from 2007 to 2018, using 30 walk-forward windows with a 36-month gap between validation and test periods and only information available when a loan is funded. A decision threshold chosen by validation return on investment (ROI) was compared with the default threshold of 0.5, lending to every loan a grade-based rule and logistic regression, with full and downsampled training data, five random seeds, and paired tests across windows. The random forest reached a mean AUC of 63.7–64.6%, close to logistic regression (64.2–65.0%). Overall windows, the ROI-based threshold raised mean window ROI from 4.04% (lend-all) to 4.97% (difference 0.92 percentage points, 95% CI −0.29 to 2.23) while lending 51% of the capital; no difference remained significant after Holm adjustment. The share of test loans resolved by the end of the data fell from over 99% in windows 1–19 to 7% in window 30, so the negative returns of late windows reflect censoring. In the 19 matured windows, lend-all earned 9.08% on average and the random forest did not improve on it: ROI-based thresholds lowered mean ROI by 0.92–1.15 percentage points and profit by 40–54%. Return-based evaluations of default models therefore need time-ordered tests that account for unresolved loans.
A Data-Driven Multi-Variable Adaptive Closed-Loop Control Framework for Real-Time Energy Efficiency Optimization Anang Prasetyo; Sudarno, Prabowo Wahyu; Dharaka, Kumara Pinasthika
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16471

Abstract

The increasing energy consumption and urbanization of the built environment necessitates better energy management solutions to combat climate change. Despite the optimal energy savings offered by modern approaches such as Reinforcement Learning (RL) and Model Predictive Control (MPC), their real-world implementation is often limited by high computational costs, complex thermal modeling, and the requirement of extensive training data. To bridge this gap, this paper proposes a data-driven, multi-variable adaptive closed-loop control architecture for real-time energy efficiency optimization in smart buildings. The system dynamically adjusts power consumption by leveraging occupancy proxies and external climatic parameters, particularly wind speed and dew point, to account for air infiltration and latent heat loads, using a lightweight decision logic. The system's operational constraints are strictly tuned to ASHRAE 55 standards to guaranty accurate thermal comfort. By numerical simulations with UCI Appliances Energy Prediction dataset, the proposed adaptive system achieves a significant 28.01% reduction in overall energy consumption, and a 26.85% reduction in peak demand compared to baseline operations. This efficiency is estimated to have reduced the system's carbon footprint by 28.01%. These findings demonstrate that the integration of multi-variable climatic data within an easily deployable, heuristic adaptive control logic represents a highly responsive and sustainable alternative to computationally intensive predictive systems, greatly improving modern building management with no compromise to occupant well-being.
2006 Yogyakarta Synthetic Time History Response Spectrum versus SNI 1726:2019: Building Response Rozaq, Abdul; Jafar, Jafar; Saputra, Elvis
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16476

Abstract

This study investigates how a response spectrum derived from a synthetic acceleration time history representing the 2006 Yogyakarta earthquake compares with the SNI 1726:2019 design spectrum and influences the seismic response of a five-story reinforced concrete building. A 5%-damped event-based spectrum was developed through linear-acceleration analysis of single-degree-of-freedom systems and applied alongside the SNI spectrum to a three-dimensional ETABS model for Site Class SD. The developed spectrum reached a peak pseudo spectral acceleration of 1.011 g at 0.34 s, exceeding the SNI plateau of 0.78 g (0.155–0.777 s) by 29.62%. At the building's fundamental period of 1.086 s in X, the developed and SNI spectra gave 0.58 g and 0.56 g, a difference of 3.51%. At 1.385 s in Y, they gave 0.32 g and 0.44 g, a difference of 31.58%. The corresponding differences in unscaled base shear, elastic roof displacement, maximum inter-story drift, and P-Delta stability coefficients were small in X (2.28–3.05%) but greater in Y (27.35–31.34%), where SNI generally produced higher demands. Force scaling brought all cases to 2961.020 kN, although story-level force distributions remained different. All drift and stability checks met the adopted limits. These findings indicate that period-dependent spectral differences are reflected in structural responses but remain specific to the synthetic input and building examined. Further studies using additional ground motions and different building configurations are recommended.
Comparison of Normalization Methods in Spectral Biclustering on Low Birth Weight Data in Gorontalo Province Panyili, Shania Rahmadani P.; Zubedi, Fahrezal
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16534

Abstract

Low Birth Weight (LBW) remains a major maternal and child health challenge associated with elevated risks of neonatal mortality and stunting. Conventional one-way clustering algorithms partition health centers or risk indicators separately, failing to identify local interactions across specific subsets of variables simultaneously. This study conducts a comparative evaluation of three normalization schemes within the Spectral Biclustering algorithm—Independent Rescaling of Rows and Columns (IRRC), Bistochastization, and Log-Interactions—to identify optimal bicluster structures on 2024 LBW data across 95 public health centers in Gorontalo Province. Preprocessing procedures addressed missing observations using mean imputation prior to variable standardization via Z-scores. The parameter tuning framework systematically evaluated 105 parameter combinations across variations of singular vectors (E = 2 to 8) and intra-bicluster variance thresholds (withinvar = 1 to 5), yielding 17 candidate models that satisfied stability criteria. Model validation using the Average Squared Residue (ASR) demonstrates that Bistochastization achieves optimal bicluster homogeneity, generating eight biclusters with the lowest ASR value of 0.1755535 at configuration E = 6 and withinvar = 3. In comparison, Log-Interactions and IRRC achieved higher ASR values of 0.3552414 (eight biclusters) and 0.3623521 (six biclusters), respectively. Structural bicluster simplification reveals distinct spatial patterns, grouping maternal health service disparities and maternal physiological risk profiles across specific public health centers. These empirical findings provide an analytical foundation for targeted maternal health interventions across regional health centers.
Development of an Extreme Rainfall-Based Hazard Index for the Air Bengkulu Watershed, Indonesia Kurniawan, Bima; Gunawan, Gusta; Misliniyati, Rena; Besperi, Besperi; Amri, Khairul
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16557

Abstract

Extreme rainfall is one of the main hydrometeorological factors that drive flood-related hazards in tropical watersheds. However, individual rainfall indicators may not sufficiently capture the combined spatial characteristics of extreme precipitation. The objective of this research is to develop a Rainfall Hazard Index (RHI) for the Air Bengkulu Watershed, Indonesia, based on multiple indices of extreme rainfall. Daily rainfall data of four stations for the period 2007-2025 are analyzed to compute Maximum 1-Day Rainfall (Rx1day), Maximum Consecutive 5-Day Rainfall (Rx5day), Very Wet Days precipitation (R95p) and Rainfall Trend (RT). Rainfall trends were assessed by Mann–Kendall test and Sen’s slope estimator. Inverse Distance Weighted (IDW) interpolation was applied to generate spatial distributions. The indicators were combined with Analytical Hierarchy Process (AHP) weighting to construct the RHI. The RHI was classified into five hazard categories by Jenks Natural Breaks. The AHP results gave the highest weight to Rx5day (0.471) followed by R95p (0.284), Rx1day (0.172) and RT (0.073) with a Consistency Ratio of 0.0185. The hazard class of extremely high rainfall covered 60.73% of the watershed, mainly in the upstream areas. A preliminary comparison with historical flood records from the Regional Disaster Management Agency (BPBD) and direct field observations revealed a match at 11 of 15 locations (73.3%). These findings show RHI as a first stage spatial screening tool for rainfall-related hazards that can be used in watershed management and spatial prioritization rather than full flood prediction.
Styrofoam Waste as Partial Pen 60/70 Binder Replacement in AC-BC Marshall Characteristics Lestari, Cinta Ayu; Bahri, Samsul; Fitriansyah, Aidil; Mawardi, Mawardi; Supriani, Fepy
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 2 (2026): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i2.16777

Abstract

Road pavements need materials that can improve asphalt mixture performance. This study analyzes styrofoam waste as a partial replacement for Pen 60/70 asphalt in Asphalt Concrete–Binder Course (AC-BC) mixtures. Styrofoam contents of 6%, 9%, and 12% by weight of asphalt (variations V1, V2, and V3) were compared with a normal mixture as the control. The styrofoam was added to hot asphalt before mixing with the aggregate (wet process). Marshall testing on three specimens per mixture covered Voids in Mineral Aggregate (VMA), Voids Filled with Asphalt (VFA), Voids in Mixture (VIM), flow, stability, Marshall Quotient (MQ), and density, following the 2025 General Specifications of the Directorate General of Highways (Direktorat Jenderal Bina Marga, 2025). The optimum asphalt content (OAC) was determined and compared with the control and specification limits. All variations met the AC-BC requirements. The OAC was 5.37% for the normal mixture and 5.12% for V1–V3, but the confirmation specimens were prepared at 5.37% (normal) and 4.87% (V1–V3), so their asphalt contents differ. Compared with the normal mixture, V1 (6% styrofoam, 94% asphalt) had higher stability (978.13 versus 896.46 kg) and MQ (303.98 versus 271.69 kg/mm), slightly lower flow (3.23 versus 3.30 mm) and VMA (14.86% versus 15.50%), and slightly higher VIM (3.94% versus 3.72%). No difference was statistically significant (ANOVA, p > 0.05), so V1 is only the most promising variation. The findings are limited to laboratory-scale Marshall results.