This paper begins by discussing the implementation of monitoring and control strategies to improve industrial performance. Industrial performance is presently defined in terms of multiple and multi-level criteria, which need to be synthesized for overall improvement purposes. The Multi-Attribute Utility Theory (MAUT) provides tools relevant to decisionmaking within this specific context. According to our approach, overall company performance is expressed as the aggregate value of each elementary performance relative to each criterion, with the aggregation operator being a Choquet integral. The paper then recalls previous studies regarding the design of efficient improvements within this same MAUT framework, before focusing on an analysis of overall performance gain in terms of elementary performance improvements. The contribution of an aggregate performance gain criterion plays a key role in this study. Two improvement strategies will moreover be examined and compared in light of this concept development. A case study has also been included to illustrate the proposals set forth in the paper.