Article Per Year (5 Year)
p-Index From 2020 - 2025
0.562
P-Index
This Author published in this journals
All Journal International Journal of Engineering, Science and Information Technology
Ar Razi, Ar Razi
Unknown Affiliation
Astronomy Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Library & Information Science Materials Science & Nanotechnology Mathematics Mechanical Engineering Physics Social Sciences Transportation

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Comparison of the Results of the Weighted Moving Average Method and the Least Absolute Shrinkage and Selection Operator Method for Predicting Total Palm Oil Production at PT. Mora Niaga Jaya Ardiansyah, Sakha; Dinata, Rozzi Kesuma; Ar Razi, Ar Razi
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.862

Abstract

This study compares two prediction methods, Weighted Moving Average (WMA) and Least Absolute Shrinkage and Selection Operator (LASSO), in forecasting the total palm oil production at PT. Mora Niaga Jaya. Accurate forecasting is essential in the palm oil industry to support decision-making, optimize production planning, and manage supply chains efficiently. The WMA method produced more realistic prediction results, with a Mean Absolute Error (MAE) of 114,854 tons and a Mean Absolute Percentage Error (MAPE) of 220.45%, despite still having a considerable margin of error. These values suggest that while WMA is not perfectly accurate, it performs moderately well, given the complexity and variability inherent in agricultural production data. On the other hand, the LASSO method yielded significantly worse results, with an extremely high and unrealistic MAE and a MAPE of 291,456.000%, indicating that this approach is unsuitable for palm oil production forecasting in this specific case. The underperformance of the LASSO method may be due to the nature of the data used, which may not meet the assumptions required for LASSO to function optimally, such as linear relationships and minimal noise. This highlights the importance of aligning forecasting methods with the dataset's characteristics. Based on the comparison, it can be concluded that the WMA method is more appropriate for predicting palm oil production than LASSO. However, further steps such as parameter optimization, data normalization, and outlier removal should be undertaken to achieve better predictive accuracy. This research provides valuable insights into the importance of selecting the correct predictive method and ensuring data quality in forecasting. Ultimately, careful model selection and data preprocessing support effective operational and strategic decisions in the palm oil industry.
Implementation of the Simple Additive Weighting Algorithm for Café Recommendations in Lhokseumawe City Arkan, Raihan; Safwandi, Safwandi; Ar Razi, Ar Razi
International Journal of Engineering, Science and Information Technology Vol 5, No 3 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i3.885

Abstract

The selection of cafés that match customer preferences is a challenge, especially in the city of Lhokseumawe, which has 30 cafés with different characteristics. This research implements the Simple Additive Weighting (SAW) algorithm to provide recommendations for the best café based on six criteria, namely price (weight 0.25), menu (0.2), order duration (0.15), service (0.2), facilities (0.15), and discounts promotions (0.05). The recommendation system was developed using a combination of Laravel PHP and Python, where Laravel is used to build an interactive web interface. Python also plays a role in data processing and complex mathematical calculations. The results showed that the system was able to provide optimal recommendations, with Petrodollar Coffeeatery Roastery as the top choice based on the calculation of the highest preference values (3.28 for price, 2.48 for menu, 3.16 for order duration, 2.88 for service, 2.96 for facilities, and 2.8 for discounts promotions). TR Coffee and Platinum Coffee occupy the following positions. In addition, this study found that the weight of the criteria and the number of datasets (150 reviewers) significantly influence the quality of recommendations. The more representative the weights used and the larger the dataset analyzed, the more accurate the system will produce recommendations based on user preferences. Thus, weight optimization and dataset expansion are essential factors in improving the effectiveness of SAW-based recommendation systems.
Analysis of Boarding House Feasibility and Satisfaction Using Data Mining with the C4.5 Algorithm Based on Service Quality and Facilities Manik, Aktina; Abdullah, Dahlan; Ar Razi, Ar Razi
International Journal of Engineering, Science and Information Technology Vol 5, No 3 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i3.896

Abstract

This research develops a boarding house eligibility classification system using the C4.5 algorithm based on service quality and available facilities. The system evaluates boarding house eligibility by considering various factors such as management services, cleanliness, security, room facilities, public facilities, internet access, comfort, and price. Each of these factors is given a specific weight based on its importance to the tenants, and they are used to classify boarding houses as luxury, standard, and economical. The classification results show that 43% of luxury boarding houses were deemed eligible, while 57% were not. In the standard boarding house category, 21% were classified as eligible, and 79% as ineligible, while in the economical category, 23% were eligible and 77% were ineligible. Using the Confusion Matrix and Classification Report, model evaluation revealed precision ranging from 0.4 to 1.0, recall from 0.67 to 1.0, and F1-scores from 0.5 to 0.91, demonstrating a reasonably high overall accuracy. Additionally, feature importance analysis revealed that price, water and electricity availability, and room facilities are the most influential factors in determining boarding house eligibility. The system's performance was tested against a dataset of real-world boarding houses, and the results suggest that it can accurately classify boarding houses based on key factors that affect tenant satisfaction. The system has the potential to serve as a valuable decision-making tool for boarding house owners, helping them improve service quality and for prospective tenants, enabling them to make more informed housing choices based on their preferences and needs.
Co-Authors Ardiansyah, Sakha Arkan, Raihan Dahlan Abdullah Manik, Aktina Rozzi Kesuma Dinata Safwandi Safwandi, Safwandi