1 INTRODUCTION Measuring corporate performance is difficult and challenging. In different decision-making contexts stakeholders tend to use different criteria and meth- odologies, thus arriving at different (and contrasting) assessments of the sustainability of corporate per- formance in practice. Let us consider methodology where the assess- ment of corporate performance is based on optimiza- tion algorithms. This includes so called eco- efficiency models, multi-attribute, and multi-criteria decision making models (Jablonský 2007). Data En- velopment Analysis (DEA) models (Charnes et al. 1978) also belong to this category. DEA is a non- parametric methodology aimed at evaluating the rel- ative efficiencies of comparable decision-making units (hereafter DMUs) by means of a variety of mathematical programming models (Charnes et al. 1978, Lee & Saen 2012). One recognized advantage of DEA is that no prior assumptions concerning the specific functional relationship linking inputs and outputs are imposed. Conventional DEA analysis al- lows the researcher to assess the performance of in- dividual DMUs taking only into account observed quantities of marketable inputs and outputs (Doyle & Green 1995, 33]. The objective of this paper is to optimize corpo- rate performance of Food Processing sector. In do- ing so, we adopt the GRI G4 methodology for Food Processing sector (G4 Sector Disclosures 2014) in- troduced in Section 3. More specifically, core envi- ronmental, social and corporate governance (ESG) and economic indicators at corporate level are using DEA techniques introduced in the Section 3. Com- puting company score by DEA using Maple (Maple 2014) is presented in Section 4. However, we go be- yond merely calculating core indicators for each in- dividual company. Instead, we aim to discover how companies score in relation to chosen simple ESG factors, a highly relevant issue in the design of poli- cy measures intended to correct corporate perfor- mance stemming from inappropriate company tech- niques. As a result, one of the main contributions of our research to previous literature in this field is the development and computation of pressure-specific composite indicators of corporate performance. Optimization of Corporate Performance Using Data Envelopment Analysis with Maple J. Hřebíček & O. Trenz Mendel University, Brno, Czech Republic Z. Chvátalová Brno University of Technology, Brno, Czech Republic J. Soukopová Masaryk University, Brno, Czech Republic ABSTRACT: Data envelopment analysis (DEA) gives a very powerful tool to decision makers in an organi- zation. DEA is then a natural choice for an application of operations research or mathematical modelling of an evaluation of corporate performance. This method was initially proposed to evaluate the efficiency. In our pa- per the level of efficiency represents the level of corporate performance. The efficiency is in our case repre- sented as a share of output in weighted sum of inputs. In other words, it represents a certain degree to which desirable outputs can offset environmental, social and corporate governance (ESG) indicators. Hence we have to consider appropriate inputs (data for calculation of ESG key performance indicators) and three outputs (complex environmental, social and corporate governance indicator). In the next stage, the fourth inputs (the chosen economic indicators) are added. Their impacts on the changing of efficiency score is under considera- tion with regression analysis, which is done by Maple. At the same time, we have observed a possible connec- tion between the achieved efficiency score and corporate sustainability. In addition to efficiency score DEA method provides weights of particular inputs and outputs. These weights are used to find those contributions of particular criteria to the achieved score. This enables us to determine the strong and weak points of the or- ganization.