Frontend Mobile Design And Water Surface Temperature Prediction Feature For Water Quality Measurement From Drone Data

Ilyas, Riski (2025) Frontend Mobile Design And Water Surface Temperature Prediction Feature For Water Quality Measurement From Drone Data. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025211189-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy

Abstract

Pencemaran air laut telah mengalami peningkatan signifikan akibat intensifikasi aktivitas manusia seperti operasi industri, limbah rumah tangga, dan pembuangan polutan. Kondisi ini berdampak serius terhadap ekosistem laut dan kelangsungan hidup biota laut, serta berpotensi mengancam kesehatan masyarakat pesisir yang bergantung pada sumber daya laut. Oleh karena itu, sistem pemantauan rutin dibutuhkan untuk deteksi dini kerusakan lingkungan dan upaya pencegahan. Penelitian ini membahas pengembangan Drones for Marine Environmental Quality (DroneMEQ), sebuah sistem aplikasi mobile yang membantu pengawas lapangan dalam memantau kualitas air secara langsung di kawasan konservasi laut. Sistem ini menggunakan drone yang dilengkapi sensor lingkungan untuk mengambil sampel dalam setiap kegiatan pemantauan. Aplikasi mobile menampilkan data kualitas air yang diambil dari sensor drone, termasuk Total Dissolved Solids (TDS), suhu, tingkat pH, dan pengukuran kekeruhan. Aplikasi DroneMEQ juga menyediakan fitur prediksi suhu menggunakan model Machine Learning Gradient Boosting. Sistem ini dilengkapi visualisasi geospasial yang memungkinkan pengelola kawasan konservasi untuk memantau aktivitas drone, melihat data pemantauan historis, dan mengakses analitik lingkungan. Selain fitur Pemantauan Aktivitas & Prediksi Suhu, terdapat pula fitur lain yang mendukung pengawas lapangan seperti Manajemen Aset, Manajemen Fasilitas, Manajemen Pelanggaran, Kondisi Cuaca, dan Rute Pengawas. Aplikasi mobile ini dikembangkan menggunakan Flutter dan menerapkan Clean Architecture. Aplikasi diuji dengan Functional Test yang menunjukkan hasil berhasil, serta menggunakan SonarQube dengan hasil evaluasi tingkat ‘A’. Evaluasi model machine learning mencapai skor R² sebesar 0,9766, RMSE sebesar 0,5499°C, dan MAE sebesar 0,2737°C. Penelitian ini menggunakan metode Gradient Boosting untuk prediksi suhu karena hasil evaluasi menunjukkan bahwa metode ini menghasilkan nilai R², RMSE, dan MAE yang lebih baik dibandingkan Random Forest, Ridge Regression, maupun CNN. Server prediksi yang dibangun menggunakan Flask dan MongoDB telah berhasil di-deploy dan diuji menggunakan Postman.
==================================================================================================================================
Marine water pollution has experienced significant increases due to intensified human activities such as industrial operations, household waste, and pollutant discharge. This condition seriously impacts marine ecosystems and the survival of marine life, while potentially threatening the health of coastal communities who depend on marine resources. Therefore, routine monitoring systems are needed for early detection of environmental damage and prevention efforts. This research discusses the development of Drones for Marine Environmental Quality (DroneMEQ), a mobile application system that helps field supervisor to monitor water quality directly in marine conservation areas. The system uses drones equipped with environmental sensors that take samples on every monitoring activity. The mobile application displays water quality data retrieved from drone sensors, including Total Dissolved Solids (TDS), Temperature, pH levels, and turbidity measurements. DroneMEQ Mobile App also provides temperature prediction feature using Gradient Boosting Machine Learning model. The system features geospatial visualization that allows conservation area managers to monitor drone activities, view historical monitoring data, and access environmental analytics. Besides Monitoring Activity & Temperature Prediction feature, there are also other features that help monitoring supervisor such as Asset Management, Facility Management, Violation Management, Weather Condition, & Supervisor Route. The mobile application was developed with Flutter and applied Clean Architecture. The Mobile Application was tested with Functional Test with successful result & SonarQube with ‘A’ overview results. The machine learning model evaluation achieved an R2 score of 0.9766, a RMSE score of 0.5499°C, and a MAE score of 0.2737°C. This research used Gradient Boosting method for temperature prediction rather than Random Forest, Ridge Regression, or CNN with Gradient Boosting since the evaluation results shows that Gradient Boosting produced better R2, RMSE, & MAE. The prediction server, built using Flask and MongoDB, has been successfully deployed and tested using Postman.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Clean Architecture, Marine Conservation Area Monitoring, Water Quality Parameters, Water Temperature Prediction
Subjects:	T Technology > T Technology (General) > T58.6 Management information systems T Technology > T Technology (General) > T58.64 Information resources management
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Riski Ilyas
Date Deposited:	28 Jul 2025 00:58
Last Modified:	28 Jul 2025 00:58
URI:	http://repository.its.ac.id/id/eprint/121664

Actions (login required)

View Item