Dataset - Machine Learning–Driven Process Optimization: A Case Study in a Poultry Farm in Southern Minas Gerais

The dataset used in the article Machine Learning–Driven Process Optimization: A Case Study in a Poultry Farm in Southern Minas Gerais contains production, management, and performance data collected from a commercial poultry farm. The variables included in the dataset are described as follows:

Dt_InRepresents the start date of the observation period during which production and management data were recorded.

Dt_FimIndicates the end date of the observation period corresponding to the collected data.

GalpãoIdentifies the poultry house or shed in which the birds were housed during the observation period.

LinhagemRefers to the genetic lineage or strain of the birds, which can influence productivity, mortality, and feed efficiency.

LoteDenotes the production batch or flock identifier, grouping birds raised together under the same management and environmental conditions.

IdadeIndicates the age of the birds at the time of data recording, typically expressed in weeks.

TotAvesRepresents the total number of birds present in the poultry house during the observation period.

MortalidadeIndicates the total number of bird deaths recorded over the specified observation period.

MortSemRepresents the number of bird deaths recorded on a weekly basis.

TotOvosSemIndicates the total number of eggs produced by the flock during a given week.

Consumo_Racao_AveRepresents the average feed consumption per bird during the observation period, typically measured in grams or kilograms per bird.

In addition to the tabular dataset, the data are also available in ARFF (Attribute-Relation File Format), enabling direct use with the WEKA data mining and machine learning software. This format facilitates reproducibility and allows researchers to readily apply classification, regression, and clustering algorithms without the need for additional data preprocessing.