Pangaribuan, Nur Indah Debora Lovely (2025) Estimasi Variabel Dan Parameter Pada Kasus Stunting Menggunakan Metode Unscented Kalman Filter. Other thesis, Institut Teknologi Sepuluh Nopember.
|
Text
5002211052-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy |
Abstract
Stunting merupakan kondisi gagal tumbuh pada anak balita akibat kekurangan gizi kronis, infeksi berulang, dan tidak mendapatkan stimulasi psikososial yang cukup. Stunting masih menjadi tantangan besar di Indonesia, dengan angka 21,5% pada tahun 2023. Beberapa daerah mungkin sudah mencapai target pemerintah Indonesia sebesar 14%, tetapi pencapaian tersebut belum merata di seluruh Indonesia sehingga banyak daerah yang memerlukan intervensi lebih lanjut untuk mencapai target nasional. Oleh karena itu, diperlukan pemantauan secara akurat terkait dinamika kasus stunting. Pada tugas akhir ini akan dilakukan estimasi variabel dan parameter menggunakan metode Unscented Kalman Filter (UKF) pada model matematika kasus stunting, dimana dalam model tersebut suatu populasi dibagi menjadi 4 kelas, yaitu anak baru lahir yang rentan stunting (S), anak bergejala stunting (E), anak stunting dan tidak bisa disembuhkan (I), dan anak bergejala stunting tetapi tidak stunting karena diberikan penanganan khusus (R). Penerapan metode Unscented Kalman Filter diharapkan dapat menghasilkan estimasi variabel dan parameter yang akurat. Penelitian ini menggunakan data tahunan prevalensi balita stunting dari tahun 2015 – 2023 yang didapatkan dari Dinas Kesehatan Kota Bekasi. Simulasi dilakukan dengan bantuan software MATLAB. Hasil simulasi menunjukkan bahwa hasil estimasi proporsi anak stunting mendekati data aktual dengan tingkat akurasi Root Mean Square Error (RMSE) sebesar 0,0003474. Dengan demikian, hasil estimasi memiliki tingkat akurasi yang tinggi.
==========================================================================================================================================
Stunting is a condition of growth failure in children under five due to chronic malnutrition, recurrent infections, and inadequate psychosocial stimulation. Stunting is still a major challenge in Indonesia, with a rate of 21,5% by 2023. Some regions may have reached the Indonesian government’s target of 14%, but the achievement has not been evenly distributed across Indonesia, so many regions need further interventions to reach the national target. Therefore, accurate monitoring is needed regarding the dynamics of stunting cases. In this final project, variable and parameter estimation will be carried out using the Unscented Kalman Filter (UKF) method in the mathematical model of stunting cases, where in the model a population is divided into 4 classes, namely newborn children who are prone to stunting (S), children with stunting symptoms (E), children who are stunted and cannot be cured (I), and children with stunting symptoms but not stunted because they are given special treatment (R). The application of the Unscented Kalman Filter method is expected to produce accurate estimates of variables and parameters. This study uses annual data on the prevalence of stunting toddlers from 2015 - 2023 obtained from the Bekasi City Health Office. Simulations were carried out with the help of MATLAB software. The simulation results show that the estimated proportion of stunted children is close to the actual data with an accuracy level of Root Mean Square Error (RMSE) of 0,0003474. Thus, the estimation results have a high level of accuracy.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Unscented Kalman Filter, Stunting, Estimasi Variabel, Estimasi Parameter, Unscented Kalman Filter, Stunting, Variable Estimation, Parameter Estimation |
| Subjects: | Q Science > QA Mathematics > QA402.3 Kalman filtering. |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Nur Indah Debora Lovely Pangaribuan |
| Date Deposited: | 01 Aug 2025 02:17 |
| Last Modified: | 01 Aug 2025 02:17 |
| URI: | http://repository.its.ac.id/id/eprint/124829 |
Actions (login required)
![]() |
View Item |
