PRIMAD-LID: A Developed Framework for Computational Reproducibility

TL;DR

PRIMAD-LID is a comprehensive framework that extends the PRIMAD model by adding modifiers to improve the control, description, and validation of computational reproducibility across disciplines.

Contribution

This work systematically integrates PRIMAD extensions into a unified, discipline-diagnostic framework with added modifiers for enhanced reproducibility practices.

Findings

Unified PRIMAD-LID framework developed

Enhanced reproducibility control with added modifiers

Applicable across diverse research disciplines

Abstract

Over the past decade alongside increased focus on computational reproducibility significant efforts have been made to define reproducibility. However, these definitions provide a textual description rather than a framework. The community has sought conceptual frameworks that identify all factors that must be controlled and described for credible computational reproducibility. The PRIMAD model was initially introduced to address inconsistencies in terminology surrounding computational reproducibility by outlining six key factors: P (Platforms), R (Research objective), I (Implementations), M (Methods), A (Actors), and D (Data). Subsequently various studies across different fields adopted the model and proposed extensions. However, these contributions remain fragmented and require systematic integration and cross-disciplinary validation. To bridge this gap and recognising that PRIMAD provides a broadly applicable framework for reproducibility in computational science, this work undertakes a focused investigation of the PRIMAD model. It combines the models previous extensions into a unified framework suitable for diverse research contexts. The result is PRIMAD-LID, a discipline-diagnostic reproducibility framework that retains the original six PRIMAD dimensions and enhances each with three overarching modifiers: Lifespan (temporal qualifier), Interpretation (contextual reasoning) and Depth (necessary granularity), thereby establishing a more cohesive and robust foundation for computational reproducibility practices.