cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
husein_ismail@uinsu.ac.id
Editorial Address
Department of Mathematic Faculty Of Science and Technology Univesitas Islam Negri Sumatera Utara IAIN St. No.1 Medan
Location
Kota medan,
Sumatera utara
INDONESIA
Zero : Jurnal Sains, Matematika, dan Terapan
Published by Universitas Islam Negeri Sumatera Utara
ISSN : 2580569X     EISSN : 25805754     DOI : 10.30829
Core Subject : Science, Education, Agriculture,
Agriculture, Biological Sciences & Forestry Chemistry Mathematics Physics
Arjuna Subject : -
Articles 258 Documents
 22   23   24   25   26 
Comparative Modeling of Pineapple Production Using Gaussian GLM and Random Forest Regression Radot MH Siahaan; Indah Gumala Andirasdini; Fuji Lestari; Dwi Mahrani; Amalia Listiani
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28721

Abstract

This study aims to conduct a comparative modelling of pineapple production at PT Great Giant Pineapple (GGP) using Gaussian GLM as parametric statistical approach and Random Forest Regression method as machine learning based on monthly data from 2014 to 2022. Multicollinearity testing and distribution fitting were conducted to validate the Gaussian assumption. For the Random Forest Regression, hyperparameters were optimized by tuning the number of trees (ntree) and the number of predictors at each split (mtry) with model stability evaluated using Out-of-Bag (OOB) error. The Gaussian GLM achieved a MAPE of 8.41% (R² = 0.106) for the GP3 clone and 11.27% (R² = 0.149) for the F180 clone. Random Forest Regression produced a testing MAPE of 9.28% (R² = 0.144) for GP3 and 12.11% (R² = 0.105) for F180. While both models achieved low prediction error based on MAPE, they differed in identifying influential variables and showed limited explanatory power as indicated by low R² values. The Gaussian GLM identifies air pressure as significant for both clones and rainfall for F180 clone, while Random Forest consistently identifies rainfall as the most influential predictor. These findings confirm the complementary strengths of parametric and machine learning approaches in supporting climate-based production planning and risk mitigation.
Geometric Data Augmentation with a Two-Stage Fine-Tuning Strategy for EfficientNetB3-Based Fruit Condition Classification Hana Sajida Azhurra; Sugiyarto Surono; Aris Thobirin
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28735

Abstract

Accurate fruit condition classification is essential for automated food safety assessment, particularly due to health risks associated with chemical contaminants such as formalin. However, reliable generalization in automated inspection systems remains challenging because limited visual variation in image datasets often leads to overfitting in deep learning models. To address this challenge, this study proposes an EfficientNetB3-based framework that integrates geometric data augmentation with a structured two-stage fine-tuning strategy to improve robustness and training stability. The proposed model achieved 99% test accuracy with consistent cross-dataset performance. The framework also demonstrated stable optimization behavior across training stages, indicating improved generalization capability. From a practical perspective, the proposed approach may support scalable food quality monitoring and automated sorting in agricultural supply chains, as well as preliminary food safety screening in large-scale inspection processes.
Optimization of Retention Levels Using the Pentikäinen Method: A Case Study on Unit-Linked Insurance Product Karin Amelia Safitri; Siti Mahsa Khalilah; Niken Yulia Astuti
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28535

Abstract

This study evaluates the adequacy of surplus reinsurance retention for unit-linked insurance products in Indonesia under changing market and regulatory conditions. Using 2024 portfolio statistics of Insurance X, the study applies the Pentikäinen method, which is grounded in ruin theory, to determine model-based retention levels and compare them with the company’s actual retention policy. The analysis incorporates premium income, claim experience, reserve considerations, and security loading assumptions. The results show that the company’s actual retention of IDR 500,000,000 is higher than the retention levels generated by the Pentikäinen model, namely IDR 275,000,000 at 0% loading, IDR 305,000,000 at 10% loading, and IDR 335,000,000 at 20% loading. Simulation results further indicate that lower retention reduces the insurer’s net retained claims and improves financial stability, as reflected in a decline in the loss ratio from 6.75% under the actual retention to 6.12%, 5.95%, and 5.74%, respectively. However, lower retention also increases reinsurance premium cessions, implying a trade-off between risk reduction and cost efficiency. The findings suggest that the retention range of IDR 305,000,000 to IDR 335,000,000 provides a more balanced outcome than both the current retention and the lowest simulated retention. This study contributes to the literature by providing an empirical application of the Pentikäinen method in surplus reinsurance for unit-linked products in Indonesia and offers a more objective basis for retention policy adjustment.
Optimization of Real-Time Object Detection in Viola-Jones Method with Enhanced AdaBoost Sucitra Sahara; Rizqi Agung Permana; Mely Mailasari
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.27876

Abstract

Face recognition is a widely used biometric technology in security systems, automated attendance, and surveillance applications. This study proposes an enhanced real-time face detection method by integrating a modified AdaBoost-based feature selection strategy into the Viola–Jones framework. The applied mathematical contribution of this study lies in formulating the optimization process as an empirical risk minimization model with adaptive boosting weight updates to reduce face recognition error. The proposed approach optimizes the weighting of weak classifiers by prioritizing Haar-like features with minimal weighted classification error at each boosting iteration, thereby improving discriminative capability. Experiments were conducted on a camera-based dataset consisting of face and non-face samples under varying illumination and pose conditions. Prior to optimization, the system achieved a precision of 70.04% and a recall of 70.05%. After applying the proposed enhancement, precision increased to 81.04% and recall to 90.02%. These results demonstrate that the modified AdaBoost integration significantly improves detection accuracy while remaining suitable for real-time face detection applications.
Identification of Subgroups of Geometric Transformations using Linear Algebra and Group Theory Ikrar Pramudya; Rubono Setiawan; Mardiyana Mardiyana; Ponco Sujatmiko; Dyah Ratri Aryuna
ZERO: Jurnal Sains, Matematika dan Terapan Vol 9, No 3 (2025): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v9i3.26564

Abstract

The set of all plane geometric transformations  forms a group under the binary operation of function composition. One of its subgroups consists of all transformations that can be expressed in the form T(x) = Ax + v, where A is an invertible (2x2) and v is a fixed vector . This study aims to identify the existence and structure of certain subgroups within through a linear algebra approach. The research methods include a literature review, simulations on specific cases to obtain a more concrete understanding of the problem, and deductive reasoning based on mathematical syllogisms to derive properties and theorems that can be algebraically verified. Consistent with the research objectives, the concepts and theoretical foundations employed are drawn from the analytical properties of plane geometry and linear algebra. These concepts and theorems are revisited to ensure their relevance to the research problem and applicability in its resolution. By applying these theoretical constructs to the problem, several subgroups whose existence can be proven algebraically are identified. These subgroups include the translation subgroup, the subgroup containing rotation transformations, the group of isometries, and the group of similarities. 
Adaptive Continuous Parameter Optimization of ARIMA using a Hybrid GA–PSO Approach for Time Series Forecasting Fitri Andini Ritonga; Sutarman Sutarman; Syahriol Sitorus
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Accurate forecasting of financial time series remains challenging due to non-stationarity, complex data patterns, and difficulties in parameter optimization within traditional models. Although ARIMA is widely used, its performance is often limited by static parameter estimation and sensitivity to evolving data structures. Existing metaheuristic-based approaches have attempted to address these issues; however, many lack adaptive mechanisms that account for varying data complexity. This study proposes a Continuous Hybrid ARIMA–Metaheuristic (GA–PSO) framework with adaptive parameter tuning guided by Model Complexity Assessment (MCA). The framework enables continuous optimization of ARIMA parameters, allowing the model to dynamically adapt to changing time-series characteristics. Empirical results demonstrate consistent improvements in forecasting performance compared to the baseline ARIMA model. For instance, in the Gold dataset (300 observations), the model achieved RMSE = 56.96, MAE = 41.86, and MAPE = 1.12%, indicating stable and accurate predictions. Statistical validation using the Diebold–Mariano test further confirms the significance of these improvements. The main contribution lies in the integration of adaptive GA–PSO optimization with complexity-aware tuning, which enhances both forecasting stability and responsiveness. However, the findings also indicate the presence of heteroscedasticity in several cases, suggesting that volatility dynamics are not fully captured by the current framework. This limitation highlights the need for incorporating volatility-aware models, such as ARIMA–GARCH, to better represent time-varying variance and improve forecasting robustness in future research.
Actuarial Evaluation of Additional Contributions in Early Retirement Programs Using the Spreading Gains and Losses Method Dwi Mahrani; Miftha Ulya Nazima; Ayu Sofia; Tiara Yulita
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28726

Abstract

This study examines the actuarial and funding implications of accelerated retirement in a defined benefit pension scheme by integrating the Projected Unit Credit (PUC) method with the Spreading Gains and Losses approach. While both methods are widely applied in pension valuation, limited empirical evidence evaluates their combined implementation under retirement age acceleration scenarios, particularly in Indonesian public sector schemes. This study addresses that gap using secondary administrative employment data of 87 female civil servants obtained from the Investment and One-Stop Integrated Services Office of Lampung Province (Dinas Penanaman Modal dan Pelayanan Terpadu Satu Pintu Provinsi Lampung), grouped into four entry-age cohorts (22–25 years). The analysis compares normal retirement at age 58 with accelerated retirement at age 50, assuming a 5% annual effective interest rate and 8% biennial salary growth. The results indicate that, at valuation age 45, actuarial liabilities increase by approximately 49.8% under retirement at age 50 relative to age 58. The shorter discounting period and earlier benefit payments outweigh the reduced contribution period, resulting in the emergence of Unfunded Actuarial Liability (UAL). The resulting Past Service Liability (PSL) is amortized over five years, requiring additional contributions ranging from IDR 27.06 million to IDR 82.05 million across entry-age groups. These findings highlight the high sensitivity of pension funding to retirement age assumptions and emphasize the importance of actuarial impact assessments prior to policy implementation. However, the deterministic framework and relatively small sample size limit broader generalization of the results.
Application of Linear Programming Based Transportation Models to Optimize Natural Disaster Relief Distribution Saddam Husein; Zulhamsyah Fachrurrazi Nasution; Ema Sri Rezeki; Afdhal Ahkrizal
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28440

Abstract

Disaster relief distribution is a complex logistical problem and requires optimal planning for targeted and efficient distribution. This study aims to apply a linear programming-based transportation model to optimize aid distribution from several warehouses to affected shelters, with clearly defined constraints based on field conditions. The method used is a quantitative approach through simulation supported by empirical data representing post-disaster conditions. The model is formulated in an objective function to minimize total distribution costs with warehouse capacity and shelter requirements constraints. The process solution model is carried out using LINDO (Linear, Interactive, Discrete Optimizer) optimization software to ensure calculation accuracy. The optimization results show a cost reduction from 1,550 to 650 units, or a savings of 58.06% while still satisfying all supply and demand constraints. These findings indicate that the linear programming-based transportation model is effective in increasing aid distribution efficiency and supporting more targeted logistics decision-making.
A Hybrid Analytical Hierarchy Process (AHP) and Profile Matching Model for E-Wallet Selection Decisions in Medan City Rahma Aulia; Ismail Husein; Rima Aprilia; Razvan Serban; Klause Roder
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28629

Abstract

The development of digital payments in Indonesia has increased the complexity of selecting an e-wallet that aligns with user preferences. This study proposes a hybrid DSS integrating AHP and Profile Matching, enhanced by a proportional transformation of AHP weights into ideal values. Unlike conventional approaches that subjectively determine ideal values, this method ensures consistency between criteria weighting and suitability evaluation, thereby reducing bias and improving ranking stability. Data from 100 students across four universities indicate that security dominates (46%), followed by convenience & access (25%), and features & cost (29%), indicating that risk reduction and trust are key adoption factors, in line with technology acceptance theory. OVO achieved the highest score. The hybrid framework reduces subjective bias in ideal-value assignment and improves ranking stability compared to standalone AHP or Profile Matching applications. These findings provide methodological contributions and practical implications for fintech providers.
Comparison of Pure Premiums for Motor Vehicle Insurance Using ZTP-Gamma GLM and Tweedie GLM Yushinta Cahya Lestari; Tiara Yulita; Amalia Listiani
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.28453

Abstract

The increasing number of motor vehicles has contributed to higher traffic density and a greater risk of accidents, thereby reinforcing the importance of protection through motor vehicle insurance. Therefore, accurately determining the pure premium is essential to maintain risk balance and ensure the sustainability of insurance companies. This study employs Generalized Linear Models, which are an extension of classical linear regression that allow the response variable to follow non-normal distributions, particularly the Zero-Truncated Poisson, Gamma, and Tweedie distributions. Using motor vehicle insurance claim data from 2022 with 386 observations, this research compares two premium modeling approaches, namely the ZTP–Gamma model for estimating claim frequency and claim severity, and the Tweedie GLM for modeling total claims in the calculation of pure premiums for motor vehicle insurance. The analysis shows that the estimated pure premiums for the ZTP–Gamma GLM range from IDR 2,138,532 to IDR 19,939,391, while the estimates for the Tweedie GLM range from IDR 2,153,665 to IDR 20,936,047. The ZTP–Gamma GLM demonstrates better accuracy, with a MAPE value of 23.65% compared to 25.844% for the Tweedie GLM, resulting in an accuracy difference of 2.194%. These findings indicate that the ZTP–Gamma GLM is more effective in producing accurate pure premium estimates.

Page 25 of 26 | Total Record : 258

 22   23   24   25   26 

Filter by Year

2017 2026


Reset
Filter By Issues
All Issue Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan Vol 9, No 3 (2025): Zero: Jurnal Sains Matematika dan Terapan Vol 9, No 2 (2025): Zero: Jurnal Sains Matematika dan Terapan Vol 9, No 1 (2025): Zero: Jurnal Sains Matematika dan Terapan Vol 8, No 2 (2024): Zero: Jurnal Sains Matematika dan Terapan Vol 8, No 1 (2024): Zero: Jurnal Sains Matematika dan Terapan Vol 7, No 2 (2023): Zero: Jurnal Sains Matematika dan Terapan Vol 7, No 1 (2023): Zero: Jurnal Sains Matematika dan Terapan Vol 6, No 2 (2022): Zero: Jurnal Sains Matematika dan Terapan Vol 6, No 1 (2022): Zero: Jurnal Sains Matematika dan Terapan Vol 5, No 2 (2021): Zero: Jurnal Sains Matematika dan Terapan Vol 5, No 1 (2021): Zero: Jurnal Sains Matematika dan Terapan Vol 4, No 2 (2020): Zero: Jurnal Sains Matematika dan Terapan Vol 4, No 1 (2020): Zero: Jurnal Sains Matematika dan Terapan Vol 3, No 2 (2019): Zero: Jurnal Sains Matematika dan Terapan Vol 3, No 1 (2019): Zero: Jurnal Sains Matematika dan Terapan Vol 2, No 2 (2018): Zero: Jurnal Sains Matematika dan Terapan Vol 2, No 1 (2018): January - June Vol 1, No 2 (2017): July - December Vol 1, No 1 (2017): January - June More Issue