Assessing whether automated forensic processes can be relied upon in practice and how they should be validated.

Digital forensic units face increasing backlogs, driven by the rapid growth of data stored on digital devices. Automation offers the potential to reduce manual effort in non-complex tasks, but only if the outputs can be relied upon in practice.

This project involved contributing subject-matter expertise on a national UK government programme to evaluate the use of automation in digital forensic processes.

The work focused on assessing command-line and API-based approaches, as well as robotic process automation techniques, and how these perform under operational conditions.

A structured validation framework was developed to test automated processes against the requirements of ISO/IEC 17025:2017 and the Forensic Science Regulator’s Codes of Practice.

This enabled the strengths and limitations of different approaches to be identified, and provided a framework that can be used by forensic units to implement and validate automation reliably.

The work highlights the importance of validating automated processes before relying on them in real-world decision-making.