Garuda - Garba Rujukan Digital

Widyo Pura Buana

Unknown Affiliation

Author-ID : 4179917

Agriculture, Biological Sciences & Forestry Humanities Computer Science & IT Economics, Econometrics & Finance Social Sciences Other

Published : 3 Documents Claim Missing Document

Claim Missing Document

Articles

1

Penggunaan Systematic Point Sample sebagai Area Master Frame dalam Mengestimasi Luas Panen Padi Hazanul Zikra; Widyo Pura Buana
Jurnal Statistika dan Aplikasinya Vol 6 No 1 (2022): Jurnal Statistika dan Aplikasinya
Publisher : Program Studi Statistika FMIPA UNJ

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21009/JSA.06105

Food data plays an important role in planning for national economic recovery. Therefore, Statistic Indonesia (BPS) developed the Area Frame Survey (AFS) in estimating paddy harvested area. AFS is an method for collecting harvested area data which was previously carried out using eye estimate. Eventhough objective sampling has been applied with a two-stage non-stratified point sampling design (point clustered by square segment), the use of a square segment 300 meters × 300 meters has the potential to generate bias due to the potential for observation points to fall outside the target population (paddy field). This study aims to analyze an alternative AFS using the EUROSTAT recommended design, namely two-phase unclustered point sampling (systematic point sampling frame). The method used is a sampling simulation using rice field population map of East Denpasar in February 2019. The simulation results show that the use of point sample frame provides a better level of accuracy and efficiency than segment sample frame.

ANALISIS PERBANDINGAN DESAIN SAMPLING SURVEI KERANGKA SAMPEL AREA (KSA) Hazanul Zikra; Widyo Pura Buana
Seminar Nasional Official Statistics Vol 2020 No 1 (2020): Seminar Nasional Official Statistics 2020
Publisher : Politeknik Statistika STIS

Badan Pusat Statistik (BPS) terus melakukan inovasi dalam rangka menghasilkan data pangan yang akurat, termasuk bekerja sama dengan Badan Pengkajian dan Penerapan Teknologi (BPPT) dalam mengembangkan metode survei KSA untuk mengestimasi luas panen padi. Meskipun sudah menerapkan kaidah sampling yang objektif, namun masih perlu dilakukan kajian terutama adanya potensi bias dari estimator survei KSA terutama untuk sawah-sawah dengan karakteristik tidak beraturan, kecil dan menyebar. Salah satu potensi penyebab bias dengan karakteristik sawah tidak beraturan adalah penggunaan segmen berbentuk persegi berukuran 300 meter × 300 meter menimbulkan potensi titik amatan tidak jatuh di sawah. Penelitian ini bertujuan menganalisis perbandingan beberapa alternatif desain sampling KSA, yaitu dot sampling method, two-phase systematic unclustered point sampling, dan irregular segment shape sampling atau kerangka sampel segmen yang mengikuti bentuk sebaran sawah. Metode yang digunakan adalah simulasi sampling dengan opsi desain sampling KSA tersebut dengan berbagai opsi jumlah sampel menggunakan peta populasi sawah di Kecamatan Denpasar Timur pada bulan Februari 2019. Perbandingan hasil simulasi dianalisis dengan pengujian beda rata-rata MSE dan optimasi biaya pada seluruh desain sampling. Kesimpulan dari penelitian ini adalah rancangan sampling menggunakan irregular segment shape dengan 3 (tiga) segmen memberikan tingkat akurasi dan efisien yang lebih baik dibandingkan rancangan sampling dengan kerangka sampel segmen berbentuk persegi.

Pemetaan Potensi Lahan Jagung Menggunakan Citra Satelit Dan Random Forest Pada Cloud computing Google Earth Engine Dwi Wahyu Triscowati; Widyo Pura Buana; Arif Handoyo Marsuhandi
Seminar Nasional Official Statistics Vol 2021 No 1 (2021): Seminar Nasional Official Statistics 2021
Publisher : Politeknik Statistika STIS

The availability of information on the potential of corn fields that is quickly updated is important to support economic recovery after covid 19. Maize mapping is a challenge in agriculture because the corn planting area does not have special characteristics such as rice fields, corn does not have a standard area map, and planting can be done in rice fields and dry forest lands. Another problem is the need for high computational resources if the mapping of maize is done directly or manually identified. In this study, mapping the potential of maize in East Java in selected districts automatically using machine learning on cloud computing google earth engine. With the use of GEE cloud computing, maize mapping can be carried out in large areas without being constrained by computer capabilities. This study uses a pixel-based Random Forest (RF) machine learning algorithm with input data from the Landsat-8, Sentinel-1 and Sentinel-2 satellites. Reference data to train the classification model using maize ASF results. The best accuracy of Machine learning results comes from the combination of Landsat-8 and Sentinel-2 with an average accuracy of 0.79. The classification model was then applied to 10 districts where the best result was Banyuwangi Regency with an accuracy of 0.89. Judging from the potential area of corn in the Banyuwangi area, the area ranges from 22,256.82 – 58,992.3 Ha based on pixels that are predicted to be corn at least 3 times/month. From the results of this study, it is evident that the use of cloud computing can perform calculations in 10 districts quickly, both in terms of model development and predictions. In addition, because it uses cloud computing, the use of satellite imagery can be utilized as soon as possible after the satellite image is published/released so that predictions of the potential of corn can be quickly and accurately generated. This study also highlights the shortcomings that occur, namely in terms of the number of samples for training data and the limitations of the algorithm used so that in the future it can be developed even better.

Title

Found 3 Documents
Search

Abstract

Abstract

Abstract

Title Search

Found 3 Documents Search

Abstract

Abstract

Abstract

Title

Found 3 Documents
Search