Dedy Kurnia Sunaryo
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

ANALISA PERBANDINGAN METODE CELLULAR AUTOMATA ANN DAN MARKOV UNTUK PREDIKSI TUTUPAN LAHAN DI KOTA BLITAR Arafah, Feny; Irenius Yopy Santrum; Dedy Kurnia Sunaryo; Hery Purwanto
Jurnal Tekno Global Vol. 13 No. 02
Publisher : UNIVERSITAS INDO GLOBAL MANDIRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36982/jtg.v13i02.4853

Abstract

ABSTRACT The development of urban areas in Blitar City, which is triggered by population growth and mobility, has caused changes in land cover, especially the reduction in rice fields due to land conversion for housing and infrastructure. As land cover changes occur significantly, it is necessary to develop methods to predict land cover, one of the methods is Cellular Automata (CA). The objectives of this study are to determine the results of land cover classification and prediction in 2024 by utilizing Sentinel 2A image data and comparing its accuracy with field data. The CA methods used are the CA ANN and CA Markov methods. To predict land cover in 2024, land cover data for 2016 and 2020 is needed. From the results of Sentinel 2A image processing for 2016, 2020, and 2024 using the supervised classification maximum likelihood method, 5 land cover classes were obtained including residential class, industrial/commerce/office building class, rice field class, garden class, and urban forest/ green belt/city park. The results of the 2024 classification show that the land cover area for the residential class reached 41.97%, industrial buildings 3.33%, rice fields 43.00%, gardens 0.44%, and urban forests/green belts 11.26%, with results Accuracy tests on field validation data produced an overall accuracy value of 96.07% and kappa of 94.92%. To test the accuracy of land cover prediction results against field validation data, the CA-ANN method produces an overall accuracy value of 74.51% and kappa 65.57%, while the CA-Markov method produces an overall accuracy value of 68.63% and kappa 57.80 %. Accuracy test of image classification results in 2024, the CA-ANN method produces an overall accuracy value of 72.54% and kappa 63.73%, while the CA-Markov method produces an overall accuracy value of 64.70% and kappa 53.40%. The accuracy test results show that the accuracy obtained by the CA-ANN method is classified as substantial suitability, while the accuracy obtained by the CA-Markov method is classified as medium suitability. This shows that in this study the CA-ANN method has better results for land cover prediction because it has a better level of accuracy than the CA-Markov method. Keywords : Cellular Automata ANN, Cellular Automata Markov, Supervised Classification, Land Cover   ABSTRAK Perkembangan wilayah perkotaan di Kota Blitar yang dipicu oleh pertumbuhan penduduk dan mobilitas menyebabkan perubahan tutupan lahan, terutama dengan berkurangnya lahan persawahan akibat alih fungsi lahan untuk pemukiman dan infrastruktur. Adanya perubahan tutupan lahan yang terjadi secara signifikan, maka diperlukan pengembangan metode untuk melakukan prediksi tutupan lahan, salah satu metodenya adalah Cellular Automata (CA). Tujuan yang ingin dicapai dari penelitian ini adalah untuk mengetahui hasil klasifikasi dan prediksi tutupan lahan tahun 2024 dengan memanfaatkan data citra Sentinel 2A serta membandingkan keakuratannya dengan data lapangan. Metode CA yang digunakan adalah metode CA ANN dan CA Markov. Untuk melakukan prediksi tutupan lahan tahun 2024 diperlukan data tutupan lahan tahun 2016 dan 2020. Dari hasil pengolahan data citra Sentinel 2A tahun 2016, 2020, dan 2024 dengan menggunakan metode klasifikasi terbimbing maximum likelihood, diperoleh 5 kelas penutupan lahan di antaranya kelas pemukiman, kelas bangunan industri/ perdagangan/perkantoran, kelas sawah, kelas kebun, dan kelas hutan kota/jalur hijau/taman kota. Hasil klasifikasi tahun 2024 menunjukkan bahwa luas untuk kelas pemukiman mencapai 41,97%, bangunan industri 3,33%, sawah 43,00%, kebun 0,44%, dan hutan kota/jalur hijau 11,26%, dengan hasil uji akurasi terhadap data validasi lapangan menghasilkan nilai Overall Accuracy sebesar 96,07% dan kappa 94,92%. Untuk uji akurasi hasil prediksi tutupan lahan terhadap data validasi lapangan, metode CA-ANN mempunyai nilai Overall Accuracy 74,51% dan kappa 65,57%, sedangkan untuk metode CA-Markov menghasilkan nilai Overall Accuracy 68,63% dan kappa 57,80%. Uji akurasi terhadap hasil klasifikasi citra tahun 2024, metode CA-ANN menghasilkan nilai overall accuracy 72,54% dan kappa 63,73%, sedangkan metode CA-Markov menghasilkan nilai overall accuracy 64,70% dan kappa 53,40%. Hasil uji akurasi tersebut menunjukkan bahwa akurasi yang diperoleh oleh metode CA-ANN tergolong ke kesesuaian substansial, sedangkan akurasi yang diperoleh oleh metode CA-Markov tergolong ke kesesuaian menengah. Hal ini menunjukkan bahwa dalam penelitian ini metode CA-ANN mempunyai hasil yang lebih baik untuk prediksi tutupan lahan karena mempunyai tingkat akurasi yang lebih baik dibandingkan metode CA-Markov. Kata Kunci : Cellular Automata ANN, Cellular Automata Markov, Supervised Classification, Tutupan Lahan
Analisis Pengembangan Lokasi Apartemen Menggunakan Analytic Hierarchy Process berbasis Sistem Informasi Geografis Dedy Kurnia Sunaryo; Endro Yuwono; Sentot Achmadi; Tiara DE; Geoghalvin Almeriq Audolfy Caecarma
Jurnal Informatika dan Teknologi Komputer (J-ICOM) Vol 6 No 1 (2025): Jurnal Informatika dan Teknologi Komputer ( J-ICOM)
Publisher : E-Jurnal Universitas Samudra

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55377/j-icom.v6i1.8838

Abstract

East Java has Malang City as the second-largest city. Common problems that occur in big cities in Indonesia are overcrowding and increasingly limited land. The solution to this problem is to develop vertical housing types such as apartments and flats. The need for new apartment development as a response to high public demand, by creating an analytical hierarchy process and a system for apartment developers, will assist apartment developers in making decisions. With AHP, you can determine the suitability of the apartment construction location with supporting parameters, namely slope maps, road network accessibility maps, existing land use maps, spatial pattern maps, distance maps to public facilities, and water availability maps. The results of the analysis of the suitability of apartment development in the city of Malang using AHP and GIS are then studied for the suitability of land use against the RTRW so that the implementation and development of apartment development in the city of Malang are according to plan. So that decision-making can be relevant and make the city of Malang an ideal city.