Pengembangan Web Asesmen Ergonomi Berdasar NMQ Menggunakan K-Means Clustering dan Flask

Kiesly, Abiyu Ramadhan (2026) Pengembangan Web Asesmen Ergonomi Berdasar NMQ Menggunakan K-Means Clustering dan Flask. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025221123-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Evaluasi risiko ergonomi terkait gangguan muskuloskeletal (Work-related Musculoskeletal Disorders / WMSDs) umumnya dilakukan secara konvensional yang bersifat subjektif dan lambat. Penelitian ini bertujuan mengembangkan ErgoFit, sistem web asesmen ergonomi objektif berbasis Nordic Musculoskeletal Questionnaire (NMQ) menggunakan algoritma K-Means Clustering. Penelitian ini memproses 919 data observasi sekunder menjadi 678 data bersih melalui preprocessing seperti penanganan pencilan dan noise menggunakan DBSCAN. Jumlah klaster optimal dievaluasi melalui Elbow Method, Silhouette Score, dan Davies-Bouldin Index (DBI). Sistem aplikasi dikembangkan menggunakan pendekatan Decoupled Architecture dengan Next.js pada sisi frontend dan Flask pada sisi backend. Hasil komputasi menunjukkan pemisahan optimal pada dua klaster (K=2), yang divalidasi dengan pencapaian Silhouette Score sebesar 0,3294 dan DBI sebesar 1,3056. Berdasarkan karakteristik centroid keluhannya, klaster secara tegas dikelompokkan menjadi risiko rendah dan risiko tinggi. Pengujian sistem web membuktikan bahwa model berhasil terintegrasi pada dengan aplikasi web dan mampu memberikan prediksi tingkat risiko ergonomi berdasarkan kelompok risiko yang telah terbentuk. Kesimpulannya, aplikasi ErgoFit dapat digunakan sebagai instrumen deteksi dini untuk memetakan risiko gangguan muskuloskeletal pada pekerja.
========================================================================================================================================
Ergonomic risk evaluation related to Work-related Musculoskeletal Disorders (WMSDs) is generally conducted using conventional methods that are subjective and slow. This research aims to develop ErgoFit, an objective ergonomic assessment web system based on the Nordic Musculoskeletal Questionnaire (NMQ) using the K-Means Clustering algorithm. This study processed 919 secondary observation data into 678 clean data through preprocessing stages, such as handling outliers and noise using DBSCAN. The optimal number of clusters was evaluated using the Elbow Method, Silhouette Score, and Davies-Bouldin Index (DBI). The application system was developed using a Decoupled Architecture approach, utilizing Next.js on the frontend and Flask on the backend. The computational results demonstrated an optimal separation into two clusters (K=2), which was validated by a Silhouette Score of 0.3294 and a DBI of 1.3056. Based on the centroid characteristics of the complaints, the clusters were distinctly classified into Low Risk and High Risk groups. Web system testing confirmed that the model successfully integrated with the web application and was able to provide predictions of ergonomic risk levels based on the established risk groups. In conclusion, the ErgoFit application can be used as an early-detection tool to assess the risk of musculoskeletal disorders among workers.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Asesmen Ergonomi, K-Means Clustering, Musculoskeletal Disorders (MSDs), Nordic Musculoskeletal Questionnaire (NMQ), Flask, Ergonomic Assessment, K-Means Clustering, Musculoskeletal Disorders (MSDs), Nordic Musculoskeletal Questionnaire (NMQ), Flask
Subjects:	T Technology > T Technology (General) > T58.6 Management information systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Abiyu Ramadhan Kiesly
Date Deposited:	27 Jul 2026 07:22
Last Modified:	27 Jul 2026 07:22
URI:	http://repository.its.ac.id/id/eprint/138091

Actions (login required)

View Item