A situational method to enable pattern-based requirement specifications as low-code/no-code software models

Main problem. Requirement patterns have emerged as a means to support the reuse and systematic production of requirement specifications, improving both efficiency and quality—typically in the form of textual documents. In Low-Code/No-Code (LCNC) software development, however, requirements are represented as software models rather than textual artefacts, still requiring effort to transform requirement specifications into software models. Although several authors have proposed approaches to reduce the software modelling effort in LCNC tools—i.e., modelling assistance approaches—the end-to-end integration between pattern–based requirement specifications and software modelling in LCNC software development remains overlooked. Solution. Therefore, this PhD Thesis introduces MARPa: a Modelling Assistance and Requirements Pattern-based situational method for LCNC software development. MARPa enables requirements engineers, stakeholders, and LCNC developers to collaboratively develop software using LCNC tools, reusing requirement patterns and leveraging modelling assistance—from pattern selection to software model creation and refinement. Following the Situational Method Engineering approach, we enable MARPa to be tailored to specific organisations and software development contexts. Validation. We validated MARPa through empirical studies aimed at evaluating MARPa’s technical and social suitability, as well as their effects on subjects’ effectiveness, efficiency, and perceptions during requirement specification using LCNC tools to analyze MARPa’s technology acceptance. Our results show that integrating requirement patterns with modelling assistance following MARPa is technically feasible and holds the potential to improve LCNC development based on efficiency and effectiveness results. Furthermore, we identified challenges resulting from our empirical validations, providing a foundation for a research agenda.