| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 1711431 | Biosystems Engineering | 2012 | 10 Pages |
The pork processing industry struggles with ineffective procurement plans due to the complexities of pig growth estimation. In this study, artificial neural network (ANN) models were applied to estimate the pig size distribution and average weight of pigs in a herd. The results indicated that the developed models provide reasonable prediction solutions with a mean absolute error (MAE) lower than 5.0 and a root-mean square error (RMSE) lower than 7.0 for pig size distribution estimation. The estimates of the average weight of finishing pigs exhibited a MAE of 2.33 and an RMSE of 3.15. Moreover, pig size distribution was applied to the problem of determining a pig procurement plan. Using the proposed concept, a proper procurement plan for each herd can be determined by improving the precision of pig size measurements. Because of its simplicity, this concept is highly applicable to the pork processing industry and offers a potentially large cost reduction.
► The pig size distribution and the average weight of pigs in a herd are estimated. ► The artificial neural network is a practical tool for pig growth estimation. ► The integration of pig size estimation and a pig procurement plan is demonstrated. ► Precision of pig size measurements enables a low inventory cost procurement plan.
