PERHITUNGAN KAJIAN KELAYAKAN FINANSIAL PROSES PASANG BARU LISTRIK DENGAN PERLUASAN JARINGAN MENGGUNAKAN MACHINE LEARNING
Dhudhik Arief Hadiyanto, Prof. Ir. Selo, S.T., M.T., M.Sc., Ph.D., IPU, ASEAN Eng. ; Dr. Ir. Guntur Dharma Putra, , S.T., M.Sc.
2025 | Tesis | S2 Teknologi Informasi
The process of installing new electricity with network expansion requires a financial feasibility study as a basis for consideration in determining investment in electricity infrastructure. This study is a crucial component in assessing the feasibility of a project based on prospective customer requests. The parameters used in the feasibility analysis include NPV (Net Present Value), IRR (Internal Rate of Return), and PP (Payback Period), which are generally used as indicators to assess the financial feasibility of a project. Currently, the calculation of these parameters is still done manually using spreadsheets, which takes a considerable amount of time and may lead to customer dissatisfaction due to delays in obtaining information about the progress of their applications. Several studies and research on financial feasibility studies and Machine Learning have been conducted, but they remain separate and unrelated to one another. Some are still limited to manual calculation methods and have not considered the use of Machine Learning. To achieve an efficient, fast, and measurable financial feasibility study, a system is needed that can perform calculations to make decisions quickly. A Machine Learning modeling system using the Supervised Learning method was chosen to implement the feasibility study to simplify assumptions about the relationships between variables and capture complex interactions between variables. This study presents a comparison of several Supervised Learning models, including Elastic Net, Ridge Regression, Linear Regression, Random Forest, Lasso Regression, XGBoost, and SVM (Support Vector Machine). The best modeling results are then used as the foundation for a web-based prototype (localhost) that performs financial feasibility study parameter calculations, displaying graphs, cumulative cash flow, and annual cash flow tables. It is hoped that this research will support the process of new electricity connections with network expansion to run more quickly and efficiently, particularly in terms of financial feasibility calculations, enabling management to draw accurate conclusions and maintain and enhance PLN's overall image as a State-Owned Enterprise (SOE) through excellent service.
Kata Kunci : Proses, kajian, kelayakan, listrik, instalasi, Machine Learning, Supervised Learning, PLN.