Tartila, Hikmata (2026) Prediksi Pemakaian dan Penentuan Reorder Point Obat Menggunakan Ridge Quantile Regression Berbasis Time Series (Studi Kasus: Puskesmas Ibuh Kota Payakumbuh). Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
22083010082.-cover.pdf Download (2MB) | Preview |
|
|
Text (Bab 1)
22083010082.-bab1.pdf Download (356kB) | Preview |
|
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
22083010082.-bab2.pdf Restricted to Repository staff only until 21 May 2028. Download (793kB) | Request a copy |
|
|
Text (Bab 3)
22083010082.-bab3.pdf Restricted to Repository staff only until 21 May 2028. Download (687kB) | Request a copy |
|
|
Text (Baba 4)
22083010082.-bab4.pdf Restricted to Repository staff only until 21 May 2028. Download (3MB) | Request a copy |
|
|
Text (Bab 5)
22083010082.-bab5.pdf Download (209kB) | Preview |
|
|
Text (Daftar Pustaka)
22083010082.-daftarpustaka.pdf Download (202kB) | Preview |
|
|
Text (Lampiran)
22083010082.-lampiran.pdf Restricted to Repository staff only until 21 May 2028. Download (313kB) | Request a copy |
Abstract
Ensuring the availability of medicines in line with service needs is a crucial aspect of maintaining the quality of healthcare services at community health centers. However, in practice, stock shortages frequently occur and disrupt service delivery, while overstocking can lead to accumulation and an increased risk of expiration. Most community health centers still rely on the Average Monthly Consumption (AMC) method for drug planning, which only calculates average monthly usage without accounting for demand fluctuations. With the advancement of quantitative approaches, prediction methods based on Ordinary Least Squares have begun to be applied to provide more systematic estimates of drug requirements. Nevertheless, OLS has limitations, as it only produces estimates of the mean and assumes normally distributed errors, making it less capable of capturing uneven demand patterns or sudden spikes. To address these limitations, this study applies Ridge Quantile Regression as a predictive method that can estimate demand at two distribution levels: the 0,5 quantile representing typical demand and the 0,9 quantile representing extreme demand conditions. The regularization parameter λ is determined using a block bootstrap approach with out-of-bag (OOB) evaluation to obtain more stable and robust estimates against data variability. The resulting quantile predictions are then integrated into the calculation of the Reorder Point (ROP) and implemented through a Graphical User Interface (GUI) to support more efficient, adaptive, and responsive operational decision-making in managing drug inventories at community health centers.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | hikmata tartila | ||||||||||||
| Date Deposited: | 21 May 2026 06:28 | ||||||||||||
| Last Modified: | 21 May 2026 06:28 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/51980 |
Actions (login required)
 |
View Item |