Benchmarking environmental and operational parameters through eco-efficiency criteria for dairy farms

Abstract

Life Cycle Assessment (LCA) is often used for the environmental evaluation of agri-food systems due to its holistic perspective. In particular, the assessment of milk production at farm level requires the evaluation of multiple dairy farms to guarantee the representativeness of the study when a regional perspective is adopted. This article shows the joint implementation of LCA and Data Envelopment Analysis (DEA) in order to avoid the formulation of an average farm, therefore preventing standard deviations associated with the use of average inventory data while attaining the characterization and benchmarking of the operational and environmental performance of dairy farms. Within this framework, 72 farms located in Galicia (NW Spain) were subject to an LCA + DEA study which led to identify those farms with an efficient operation. Furthermore, target input consumption levels were benchmarked for each inefficient farm, and the corresponding target environmental impacts were calculated so that eco-efficiency criteria were verified. Thus, average reductions of up to 38% were found for input consumption levels, leading to impact reductions above 20% for every environmental impact category. Finally, the economic savings arising from efficient farming practices were also estimated. Economic savings of up to 0.13 € per liter of raw milk were calculated, which means extra profits of up to 40% of the final raw milk price.

Introduction

The recognition of the food sector as a relevant source of environmental impact demands the evaluation of food production systems in a comprehensive way (Garnett, 2008, Mattsson and Sonesson, 2003). In particular, livestock-related systems are considered a major environmental issue (Garnett, 2009, Steinfeld et al., 2006). In this context, Life Cycle Assessment (LCA) has been widely used to evaluate the environmental impact of livestock products (de Vries and de Boer, 2010, IDF, 2009). Pork, chicken, beef, milk and eggs are among the most frequently assessed food products.

LCA is an internationally standardized methodology (ISO, 2006a, ISO, 2006b) to assess the environmental aspects and potential impacts associated with a product by means of (i) the compilation of an inventory of relevant inputs and outputs of a product system, (ii) the evaluation of the potential environmental impacts linked to those inputs and outputs, and (iii) the interpretation of the results of the inventory analysis and impact assessment phases in relation to the objectives of the study.

To date, several LCA studies have focused on the environmental performance of conventional and organic milk production systems at farm level. The number of farms assessed in each study is highly variable ranging from only 1 (Cederberg and Mattsson, 2000) to 119 dairy farms (Thomassen et al., 2009). On the one hand, studies dealing with a short number of farms acknowledge the need of more reliable inventories for the dairy farming stage if a global view of the dairy farming sector is pursued (Hospido et al., 2003). On the other hand, when managing multiple farms, the common procedure relies either on the definition of average inventory data (Basset-Mens et al., 2009a, Casey and Holden, 2005a) or on the report of average environmental impacts (Thomassen et al., 2009), sometimes making a rough distinction according to production rate ranges (Cederberg and Flysjö, 2004). In any case, the use of average values entails the handling of standard deviations and, therefore, data quality concerns (Reap et al., 2008a, Weidema and Wesnaes, 1996). In this respect, data variability and uncertainty constitutes a challenging issue in the environmental assessment of agri-food products (Basset-Mens et al., 2009b).

In a scenario where multiple input/output data are available for multiple facilities, Vázquez-Rowe et al. (2010) recommend the use of a methodological approach based on the joint implementation of LCA and Data Envelopment Analysis (DEA), which is known as LCA + DEA methodology. DEA is a linear programming methodology used to quantify in an empirical manner the comparative productive efficiency of multiple similar entities (Cooper et al., 2007). Each homogenous entity whose input/output conversion undergoes assessment is named Decision Making Unit (DMU). LCA + DEA methodology synergically combines LCA and DEA in order to provide a tool for the operational and environmental assessment of multiple DMUs (Iribarren, 2010). This approach allows eco-efficiency verification while avoiding the use of average inventory data (Lozano et al., 2009, Lozano et al., 2010, Vázquez-Rowe et al., 2010, Vázquez-Rowe et al., in press).

In this article, the eco-efficiency concept is managed in accordance with the definition from the World Business Council for Sustainable Development. Hence, eco-efficiency refers to the delivery of competitively priced goods that satisfy human needs while progressively reducing environmental impacts of goods and resource intensity throughout the entire life cycle to a level at least in line with the Earth's estimated carrying capacity (Schmidheiny, 1992).

Iribarren et al. (2010) discussed a set of unexplored potentials of LCA + DEA methodology. Interestingly, the performance of super-efficiency analyses to identify best performing units from an operational perspective was highlighted along with the applicability of LCA + DEA methods to attain economic enhancement of LCA results, therefore improving the use of LCA to support decision making (Reap et al., 2008b).

This article uses LCA + DEA methodology with the aim of performing an eco-efficiency assessment of a high number of dairy farms. Additionally, a super-efficiency study is undertaken in order to suggest the best performing farms. Furthermore, the economic gains linked to the change from inefficient to efficient farming practices are also estimated.