PREDICTION OF THE NUMBER OF BLOOD DONORS USING XGBOOST BASED ON HISTORICAL AND DEMOGRAPHIC DATA ANALYSIS (CASE STUDY UDD PMI BOJONEGORO REGENCY)

Saputra, Moch Dani Ferdian (2026) PREDICTION OF THE NUMBER OF BLOOD DONORS USING XGBOOST BASED ON HISTORICAL AND DEMOGRAPHIC DATA ANALYSIS (CASE STUDY UDD PMI BOJONEGORO REGENCY). Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 22081010147. -cover.pdf Download (789kB)
	Text (Bab 1) 22081010147. -bab1.pdf Download (90kB)
	Text (Bab 2) 22081010147. -bab2.pdf Restricted to Repository staff only until 25 June 2029. Download (365kB)
	Text (Bab 3) 22081010147. -bab3.pdf Restricted to Repository staff only until 25 June 2029. Download (1MB)
	Text (Bab 4) 22081010147. -bab4.pdf Restricted to Repository staff only until 25 June 2029. Download (5MB)
	Text (Bab 5) 22081010147. -bab5.pdf Download (11kB)
	Text (Daftar Pustaka) 22081010147. -daftarpustaka.pdf Download (92kB)
	Text (Lampiran) 22081010147. -lampiran.pdf Restricted to Repository staff only until 25 June 2029. Download (486kB)

Abstract

Blood supply chain management faces significant challenges in maintaining a balance between stochastic supply and uncertain demand. Supply uncertainty due to various factors, including the impact of the COVID-19 pandemic, highlights the need for an accurate prediction system to optimize blood inventory. This study aims to develop and evaluate a prediction model for the daily and monthly number of blood donors at the Blood Donor Unit (UDD) of the Indonesian Red Cross (PMI) in Bojonegoro Regency using the Extreme Gradient Boosting (XGBoost) algorithm. The data used in the modeling includes 146,447 historical and demographic donor records from January 2019 to December 2024. To achieve better accuracy, this study developed three separate prediction models based on demographic segmentation, namely models based on blood type, gender, and age group. The performance of each model was evaluated at both the daily and monthly aggregate levels using three main metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the Coefficient of Determination (R²). The test results show that the XGBoost algorithm is capable of providing excellent prediction performance across all scenarios. In daily scale predictions, the model successfully captured the fluctuations in the number of donors with R² values ranging from 0.755 to 0.974. On a monthly scale, the model's performance increased significantly with R² values reaching 0.952 to 0.997. This accuracy improvement proves that the monthly data aggregation process effectively reduces daily fluctuations, resulting in a more stable historical donation pattern. Among the three demographic scenarios tested, the age group-based model, particularly for the 17-24 years age range, proved to provide the most stable and accurate prediction performance. This study concludes that the XGBoost algorithm is highly robust and adaptive in handling the volatility of time-series data on blood donation. It is hoped that this prediction model can serve as a strategic tool for the UDD PMI to proactively plan donor recruitment, optimize inventory management, and mitigate the risks of blood shortages as well as stock wastage.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Anggraeny, Fetty Tri	NIDN0711028201	fettyanggraeny.if@upnjatim.ac.id
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Unnamed user with email 22081010147@student.upnjatim.ac.id

Date Deposited:

25 Jun 2026 04:26

Last Modified:

25 Jun 2026 04:31

URI:

https://repository.upnjatim.ac.id/id/eprint/54196