Identifikasi Kelompok Pelanggan Rumah Tangga Bersubsidi Berdasarkan Pola Konsumsi Air Menggunakan Mixed-Effects Model dan Cluster Analysis

Maharani, Nur Ismi (2026) Identifikasi Kelompok Pelanggan Rumah Tangga Bersubsidi Berdasarkan Pola Konsumsi Air Menggunakan Mixed-Effects Model dan Cluster Analysis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221037-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Air bersih merupakan kebutuhan dasar masyarakat yang penyediaannya dikelola oleh PDAM melalui sistem klasifikasi tarif pelanggan. Pelanggan rumah tangga kode tarif 1.2 di PDAM Surya Sembada Kota Surabaya merupakan kelompok penerima subsidi penuh yang memiliki karakteristik pemakaian air yang beragam. Penelitian ini bertujuan mengevaluasi pemakaian air pelanggan berdasarkan karakteristik individual dan hasil pengelompokan menggunakan Linear Mixed-Effects Model dan K Means Clustering. Data yang digunakan berupa pemakaian air bulanan 1.502 pelanggan selama periode Januari 2023 hingga September 2025. Pemakaian air dimodelkan menggunakan Linear Mixed-Effects Model dengan fixed effect berupa tren dan musiman serta random effect berupa random intercept dan random slope. Nilai random intercept dan random slope selanjutnya digunakan sebagai variabel pengelompokan menggunakan K-Means, dengan jumlah cluster optimum ditentukan menggunakan Elbow Method. Hasil penelitian menghasilkan tiga kelompok, yaitu Typical Usage sebanyak 1.468 pelanggan (97,736%), High Usage sebanyak 29 pelanggan (1,931%), dan Very High Usage sebanyak 5 pelanggan (0,333%). Kelompok Typical Usage memiliki rata-rata pemakaian 25,60 m³ dengan karakteristik pemakaian relatif rendah dan stabil. Kelompok High Usage memiliki rata-rata pemakaian 846 m³ dengan tingkat pemakaian lebih tinggi dan kecenderungan meningkat, sedangkan Very High Usage memiliki rata-rata pemakaian 2.140 m³ dan menunjukkan tingkat pemakaian tertinggi. Hasil pemetaan menunjukkan bahwa pelanggan High Usage dan Very High Usage tersebar pada beberapa wilayah pelayanan. Hasil evaluasi ini dapat menjadi informasi pendukung bagi PDAM dalam melakukan pemantauan dan verifikasi pelanggan berdasarkan karakteristik pemakaian air.
======================================================================================================================================
Clean water is a basic necessity whose provision is managed by water utilities through a customer tariff classification system. Household customers under tariff code 1.2 at PDAM Surya Sembada Kota Surabaya are fully subsidized customers with diverse water consumption characteristics. This study aims to evaluate customer water consumption based on individual characteristics and clustering results using a Linear Mixed-Effects Model and K-Means Clustering. The data consist of monthly water consumption records from 1.502 customers observed from January 2023 to September 2025. Water consumption was modeled using a Linear Mixed-Effects Model with trend and seasonal components as fixed effects and random intercepts and random slopes as random effects. The estimated random intercepts and random slopes were subsequently used as input variables for K-Means Clustering, with the optimal number of clusters determined using the Elbow Method. The results identified three customer groups: Typical Usage, consisting of 1.468 customers (97,736%); High Usage, consisting of 29 customers (1,931%); and Very High Usage, consisting of 5 customers (0,333%). The Typical Usage group had an average water consumption of 25,60 m³ and exhibited relatively low and stable consumption characteristics. The High Usage group had an average consumption of 846 m³, characterized by higher consumption levels and an increasing tendency, while the Very High Usage group had an average consumption of 2.140 m³ and exhibited the highest consumption level among the three groups. Spatial mapping showed that customers in the High Usage and Very High Usage groups were distributed across several service areas. These evaluation results can provide supporting information for PDAM in monitoring and verifying customers based on their water consumption characteristics.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Data Longitudinal, Evaluasi Pemakaian Air, K-Means, Clustering, Linear, Mixed-Effects Model, Pemakaian Air, K-Means Clustering, Linear Mixed-Effects Model, Longitudinal Data, Water Consumption, Water Consumption Evaluation.
Subjects:	H Social Sciences > HA Statistics > HA31.7 Estimation Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Nur Ismi Maharani
Date Deposited:	01 Aug 2026 06:25
Last Modified:	01 Aug 2026 06:25
URI:	http://repository.its.ac.id/id/eprint/141313

Actions (login required)

View Item