Integrating Model-Based Systems Engineering for Enhanced Digital Forensics in Crash Investigations

Traffic crashes have been an epidemic in modern life, sometimes with tragic consequences. Traffic safety initiatives have included the use of vehicle event data recorders (EDRs) to help investigators understand what happened before and during a crash event. This paper presents a model-based systems engineering (MBSE) approach aimed at augmenting the digital forensics associated with EDRs in traffic crash investigations. The goal of the modeling effort is to develop system and subsystem requirements to address the needs of the crash investigator and the courts, each as stakeholders. The MBSE approach utilized the MagicGrid V2 method to work through the problem definition with the four pillars of MBSE: requirements, structure, behavior, and parametrics. The modeling effort shows some functionality is driven by compliance with existing and upcoming regulations. Also, data attributed to the identity of the driver may be present, which means privacy concerns exist. Analysis of these requirements reveals potential conflict, which means EDR designers may have to pick one requirement or another to satisfy. The modeling effort also raises the need for privacy and security controls with a compartmentalized architecture to balance the needs for the availability of forensic data with the need to protect user information and manufacturer intellectual property. The application of the MagicGrid modeling methodology sets up the detailed design efforts for a future event data recorder. Forensic models were created for use in developing forensic systems requirements. These models follow the approach an investigator would follow in obtaining data from a crash event, while also highlighting what data is available versus what data could be considered as personally identifiable information (PII) which could be a privacy concern.