Ekasakti Jurnal Penelitian dan Pengabdian (Jun 2024)

ANALISIS KLASTERISASI DAN PERAMALAN HASIL KOMODITAS TANAMAN CABAI MENGGUNAKAN METODE K-MEDOIDS DAN FUZZY TIMES SERIES DI KABUPATEN LIMA PULUH KOTA

  • Desvina Yulisda,
  • Muthmainnah Muthmainnah,
  • Hisyam Maulana Hadi

DOI
https://doi.org/10.31933/ejpp.v4i2.1081
Journal volume & issue
Vol. 4, no. 2

Abstract

Read online

Chilli is one type of vegetable that is quite important in Indonesia, both as a commodity consumed domestically and as an export commodity. Lima Puluh Kota Regency is the centre of horticultural commodities in West Sumatra Province which continues to strive for development, especially in fruit and vegetable commodities in areas that are production centres in this area. Clustering is needed in this area because many sub-districts in Lima Puluh Kota Regency produce chilli commodities so that it can facilitate the government in determining which areas have the potential to increase and optimise the yield of chilli commodities. This research uses Cluster Analysis with the K-Medoids Algorithm method and forecasting using the Fuzzy Time Series method. The results obtained in clustering using the K-Medoids Algorithm on large chilli plants are Cluster 1; 2 sub-districts, namely Pangkalan District and Kapur IX District. Cluster 2; 7 sub-districts namely Gunuang Omeh District, Suliki District, Guguak District, Mungka District, Akabiluru District, Badger District, and Harau District. and Cluster 3; 4 sub-districts namely Payakumbuh District, Bukik Barisan District, Situjuah Limo Nagari District and Lareh Sago Halaban District. The clustering results on cayenne pepper are Cluster 1; 5 sub-districts, namely Suliki District, Mungka District, Lareh Sago Halaban District, Pangkalan District and Kapur IX District. Cluster 2; 5 sub-districts namely Guguak District, Akabiluru District, Luak District, Situjuah Limo Nagari District and Harau District, and Cluster 3; 3 sub-districts namely Gunuang Omeh District, Bukit Barisan District and Payakumbuh District. The forecasting results based on the Fuzzy Time Series algorithm for large chilli in the following year are 14313.4 Ton/Ha with a MAPE value of 37% and are 4367.8 Ton/Ha for cayenne pepper by 34% of MAPE value.

Keywords