Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Jurnal Algoritma

Analisis Perbandingan Efektivitas Klasterisasi K-Means dan Pengambilan Keputusan Topsis Melalui Pendekatan Anova Juraizah, Nadiah; Ariesta, Atik
Jurnal Algoritma Vol 22 No 1 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-1.2172

Abstract

PT. Saida Indra Panca is a company that manages outsourced labor, such as cleaning services, room attendants, and landscape maintenance. The company often faces difficulties in identifying employees eligible for training and development. Therefore, this study aims to compare the effectiveness of decision-making with and without clustering. The clustering method uses the K-Means algorithm, while the decision-making method uses TOPSIS. The research adopts the CRISP-DM approach, which includes business understanding, data collection, data preparation, modeling, evaluation, and deployment. Evaluation was conducted using ANOVA to compare the variance values of two groups: the first group with clustering and TOPSIS, and the second group with TOPSIS only. The evaluation resulted in an F-value of 5.553025 and a P-value < 0.05, indicating a significant difference between the group means. The study shows that the combination of K-Means and TOPSIS is superior to using TOPSIS alone, as it results in a more structured, efficient, and accurate decision-making process. Clustering helps group employee data based on specific characteristics, making the evaluation and ranking process more targeted. As a result, the company can improve HR management efficiency by up to 25% and enhance the accuracy in selecting employees for training. This approach provides deeper insights for developing effective data-driven HR strategies and supports better decision-making in employee management.
Monte Carlo Simulation for Seasonal Stock Prediction of Seasoning at AH FOOD Nurhalizah, Ammanda Putri; Ariesta, Atik
Jurnal Algoritma Vol 22 No 1 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-1.2191

Abstract

AH FOOD is a business operating in the food industry, selling various seasoning products such as balado seasoning, balado chili sauce, Miwon, and others. The problem faced involves seasoning products that have specific characteristics, such as a limited shelf life and dependence on the availability of raw materials. This availability is often influenced by general market conditions, weather or seasons, and raw material prices. Therefore, this study aims to predict sales stock at AH FOOD based on Indonesia's seasons, namely the rainy and dry seasons. The method used in this research to predict stock is the Monte Carlo method. This method was chosen due to its ability to handle uncertainty and seasonal variability, making it superior to other methods such as time series regression in predicting seasonal stock. The Mean Absolute Percentage Error (MAPE) was used to measure the accuracy level of the prediction simulation. The results of the accuracy using MAPE showed that the Monte Carlo method is adequate and feasible to use, with an average error value of 26% for the rainy season and 27% for the dry season. This helps AH FOOD optimize stock management, reduce losses due to product expiration, and increase storage efficiency. Based on the MAPE results, the Monte Carlo method is effectively used to predict seasoning stock sales at AH FOOD based on Indonesia's seasonal divisions.