ARIMA Prognostic Application to Bull Services for Resource Usage Optimization
| Autor: | Jude. B. Rola, Ivy Fe M. Lopez, Cherry Lyn C. Sta. Romana, Larmie S. Feliscuzo |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
biology
Operations research Computer science 020209 energy 020208 electrical & electronic engineering 02 engineering and technology biology.organism_classification Development plan Resource (project management) Mean absolute percentage error Moving average Carabao 0202 electrical engineering electronic engineering information engineering Autoregressive integrated moving average Time series Test data |
| Zdroj: | MAPR |
| DOI: | 10.1109/mapr.2019.8743539 |
| Popis: | Resource optimization is one of the keys in achieving established organization objectives. In making concrete goals, forecasting can be used. Forecasting the near future’s number of bull services (i.e. natural mating with a male purebred dairy-type carabao or technically known as carabull) may help the Philippine Carabao Center-Visayas State University (PCC-VSU) in its resource (e.g. manpower, financial, animal) target setting. This can assist in formulating livestock development plan in the genetic improvement program of PCC-VSU necessary to uplift the Philippine carabao industry; thus, can help to improve the lives of the farmers especially the smallholder carabao raisers. This study implemented a bull services Auto-Regressive Integrated Moving Average (ARIMA) timeseries forecasting which is one of the advanced forecasting models used to predict trend for the next few years. An appropriate ARIMA model was selected from the analyzed bull services data from 2002 to 2014. Data from 2015 to 2017 were utilized as testing data and bull services for 2019 to 2021 were predicted as well. The results show that the ARIMA forecast are better than the existing annual target setting method used by the center. The model predicted values are generally closer to the actual number of bull services than the center’s targeted values. In terms of Mean Absolute Percent Error (MAPE), ARIMA scored 3.31. The development of a prognostic application to predict calf or young carabao gender is recommended, if applicable. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|