Predicting Customer Numbers at PT PLN (Persero) West Nusa Tenggara Regional Main Unit Using the Prophet Time Series Model
Rida Alkausar Hardi, Rio Satriyantara, Yunita Septriana Anwar, I Gede Adhitya Wisnu Wardhana
Source record
Source: Crossref
Published: Sep 1, 2026
DOI: 10.20473/conmatha.v8i2.88166
Open original source ↗Source abstract
Electricity is a fundamental need for modern society, and in developing regions such as West Nusa Tenggara, the continuous growth in the number of customers requires PT PLN (Persero) to conduct effective resource planning to prevent potential energy crises. This study aims to predict the growth of PLN’s customer numbers using the Facebook Prophet time series model. A quantitative approach was applied using monthly customer data from PT PLN (Persero) covering the period from January 2019 to December 2024. The model was optimized through a hyperparameter tuning process, and its performance was evaluated using the Mean Absolute Percentage Error (MAPE) metric. The results demonstrate that the optimized model achieved a MAPE of 0.27% during cross-validation. Analysis of the results indicates that the model effectively captured long-term growth trends and seasonal fluctuations. These findings suggest that the Prophet model can serve as a technical reference for forecasting customer numbers, potentially supporting strategic decision-making and resource allocation at PT PLN (Persero).
Evidence graph
No public relationships recorded yet.
Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.