Date of Publication :20th November 2023

Abstract: The study aimed to forecast daily electric consumption for the Cotabato Electric Cooperative (COTELCO) using ARIMA-ANFIS Algorithm. Spanning the years 2017 to 2022, the study examined data patterns from various components of electric consumption, including commercial and residential establishments, industrial facilities, streetlights, and public buildings. By employing a non-experimental quantitative approach, the research employed mathematical modeling to predict future consumption trends. The study developed and refined an ARIMA-ANFIS hybrid model, leveraging historical data to enhance predictive accuracy. An array of metric criteria, such as R², AIC, BIC, & MAPE, is utilized to assess the model's accuracy and goodness of fit. The results indicated that the proposed ARIMA-ANFIS model outperforms prior iterations with significantly lower evaluation metric values. The ARIMA-ANFIS hybrid model's forecasting for the next 36 months offers a valuable glimpse into the expected trends and patterns in COTELCO's electric consumption. By considering the forecasted values and their associated prediction intervals, stakeholders can make informed decisions and develop strategies aligned with the projected energy consumption trajectory. The policy recommendation is proposed based on the findings of the study. Furthermore, future researchers may utilize the proposed model by using the data in another setting to confirm its predictive ability.

Reference :