Assessment of feed and economic efficiency of dairy farms based on multivariate aggregation of partial indicators measured on field

摘要

Many of the metrics used to evaluate farm performanceare only partial indicators of farm operations,which are assumed to be best predictors of the wholefarm efficiency. The main objective of this work wasto identify aggregated multiple indexes of profitabilityusing common partial indicators that are routinelyavailable from individual farms to better support theshort-term decision-making processes of the cattlefeedingprocess. Data were collected from face-to-faceinterviews with farmers from 90 dairy farms in Italyand used to calculate 16 partial indicators that coveredalmost all indicators currently used to target feedingand economic efficiency in dairy farms. These partialindicators described feed efficiency, energy utilization,feed costs, milk-to-feed price ratio, income over feedcosts, income equal feed cost, money-corrected milk,and bargaining power for feed costs. Calculations offeeding costs were based on lactating cows or the wholeherd, and income from milk deliveries was determinedwith or without considering the milk quality payment.Multivariate factor analysis was then applied to the16 partial indicators to determine simplified and latentstructures. The results indicated that 5 factorsexplained 70% of the variability. Each of the originalpartial indicator was associated with all factors in differentproportions, as indicated by loading scores fromthe multivariate factor analysis. Based on the loadingscores, we labeled these 5 factors as “economic efficiency,”“energy utilization,” “break-even point,” “milkto-feed price,” and “bargaining power of the farm,” indecreasing order of explained communality. The first3 factors shared 83% of the total communality. Feedefficiency was similarly associated with factor 1 (53%loading) and factor 2 (66% loading). Only factor 4 wassignificantly affected by farm location. Milk productionand herd size had significant effects on factor 1and factor 2. Our multivariate approach eliminatedthe problem of multicollinearity of partial indicators,providing simple and effective descriptions of farmfeeding economics. The proposed method allowed theevaluation, benchmarking, and ranking of dairy herdperformance at the level of single farms and at territoriallevel with high opportunity to be used or replicatedin other areas.