Sustainable software engineering : a literature review of evaluation

Abstract

This thesis presents a structured literature review of evaluation methods and frameworks in Sustainable Software Engineering (SSE). Following the PRISMA methodology and guidelines established by Kitchenham and Charters (2007), the review synthesises peer-reviewed literature published primarily between 2011 and 2025 across four major academic databases: IEEE Xplore, ACM Digital Library, Scopus, and Google Scholar. Three research questions are addressed: how sustainability is conceptualised across the SSE literature, what evaluation methods and frameworks exist for assessing SSE practices, and what limitations and gaps characterise current approaches.

The review finds that sustainability in SSE has evolved from a predominantly environmental and product-focused framing towards multi-dimensional frameworks encompassing technical, economic, social, individual, and process dimensions. Four categories of evaluation method are identified: energy-focused measurement, architectural evaluation approaches, integrated multi-dimensional frameworks, and process-level assessment methods. The LAGSSE framework (Zada et al., 2025) represents the most comprehensive integrated approach, while the stratified and multisystemic model of McGuire et al. (2023) offers the most theoretically advanced conceptualisation.

Six primary gaps are identified in current SSE evaluation: lack of empirical validation through real-world deployment studies; fragmentation of evaluation across dimensions; insufficient tool support at the application development level; terminological inconsistency; limited evaluation of process-level sustainability; and the absence of longitudinal studies. The thesis concludes that while the conceptual foundations of SSE evaluation are well established, the practical infrastructure for validated, integrated, and tool-supported assessment remains underdeveloped relative to the ambition of the field.