68 / 2025-04-15 16:28:50

Quantifying environmental non-uniformity in buildings via revised Gini coefficient

Indoor environment, Coefficient of variation, Gini coefficient, Dimensionless normalization, Broiler building, CFD

1. 畜禽集约化养殖与可持续发展 > （1）大型养殖场的环境影响

Indoor environmental parameter distribution uniformity is commonly evaluated using the Coefficient of Variation (CV), calculated as the ratio of standard deviation to mean. However, the CV range spans from negative to positive infinity (i.e., [-∞, +∞]), lacks standardized assessment criteria, and becomes problematic when the mean approaches zero. Specifically, in datasets that include both positive and negative values. There is no unified comparison standard for CV. Furthermore, CV cannot provide a consistent criterion to simultaneously evaluate environmental parameters that include both interval and ratio variables. To address these limitations, a new index based on the Gini coefficient, i.e., the revised Gini coefficient (Gini_rev), is employed through the Min-Max normalization to enhance the evaluation of non-uniformity. In this study, Computational Fluid Dynamics (CFD) simulations are conducted to qualitatively and quantitatively analyze the velocity, temperature, and CO₂ concentration in a commercial broiler building, under two ventilation systems: cross ventilation and tunnel ventilation. The results revealed the following: (1) Gini_rev provides clear boundaries (i.e., [0, 1]) in the calculation results. The maximum value obtained is 0.604 for Gini_rev of velocity, while 2.105 for CV. (2) Gini_rev eliminates dimensional influence, which makes the distribution uniformity between different environmental parameters more equitable, resulting in a more consistent range (0.084-0.604) compared to CV's wider span (0.049-2.105). (3)During transitional seasons, cross-ventilated environments demonstrated superior uniformity, with tunnel ventilation showing 36% and 25% higher Gini_rev values for temperature and CO₂ concentration, respectively. In conclusion, Gini_rev provides a reliable method for evaluating and comparing multiple indoor environmental parameters.