Abstract: The cultivation of aquatic species in aquaculture aims at improving fish production, making the activity more dependent on human capacity, especially with what concerns to the application of efficient methods for its management. The weight and standard length are the biometric parameters mostly used to control the growth of tilapia in the cultivation process. In this work there is presented a statistical method based on transfer function ARIMA model to adjust and predict the weight of tilapia with the support of the values of the standard length. The model was adjusted to the data recorded in 21 weeks in the tilapia growing tank of the company Aquapesca de Moçambique, and the results showed that there was a good adjustment (with an explained variability of 97.8%) with which the model can be used to predict the future values of tilapia weight and make the activity more controllable and profitable.
Keywords: ARIMA model, Transfer function, predict, Aquaculture, Tilapia
Recieved: 02.05.2018 Accepted: 24.08.2018 UDC: 626.885