A MATHEMATICS researcher and university instructor proposes for the use of autoregressive integrated moving average (Arima), a mathematical model used for time series analysis, in order to forecast short-term electricity load needed per area.
In one of the lectures at the Knowledge x Change inaugural lecture series held at the University of the Philippines Mindanao on Wednesday, August 30, researcher and mathematician Kenneth Montajes said the Arima model can help forecast load demand quantity in a more accurate way.
“In my study, Arima is the backbone of the application to be used by electric cooperatives and companies. Some electric coops now use Excel in determining the electric load for the certain amount of time. It is a template and the backbone of that is Arima model. The Excel will then be a formulated model or template thru this Arima,” said Montajes.
As for his research, he studied the process specifically of the Davao Oriental Electric Cooperative (Doreco) and found out that what they are currently using still is the regression method in determining electricity load demand for their area.
“It’s stated that the results using Arima model is much better and accurate compared to the ones being used by some electric cooperatives in the region now such as Dureco,” he said.
When asked how he thinks the Arima model can help with the electric cooperatives in the event that the knowledge would be passed on to them, Montajes said thru this the electric coops will be able to balance the electrical supply and demand more efficiently.
“There will be less incurrence of penalties regarding demanding load especially that the forecast will be better and much nearer to the actual demand. Also there would be less instances of blackouts,” said Montajes.
Arima model is often used in statistics and econometrics specifically to determine time series. This can also be used in sales and trade in determining sales forecast, logistics supply and demand, and stock exchange fluctuations.
Montajes study is still ongoing but he said he would be glad to share whatever knowledge he would gain from his research to electric cooperatives and companies to help better their forecasting performances.