Integrating the IDRACS Risk Classification Tool with Georgia DCS's Case Management System

• In Event: Assessing Re-arrest Using Predictive Modeling: The Integrated Dynamic Risk Assessment for Community Supervision Tool

Abstract

The deployment of AI or advanced analytical techniques in corrections is crucial for enhancing risk assessments and optimizing resource allocation. But success may vary greatly depending on implementation. The IDRACS tool was developed via an NIJ grant for the Georgia Department of Community Supervision’s (DCS). Integration into DCS’s case management system was a coordinated effort involving researchers along with DCS IT and operations staff. This presentation describes the integration process to ensure that the tool would function seamlessly within existing infrastructures, providing accurate and actionable risk assessments. The presentation describes reviewing measures and outcomes and carrying out simultaneous data management and analytical comparison tasks to ensure that tool inputs resulted in accurate outputs for risk classification. Additionally, focus groups with community supervision officers also provided insights, leading to revised measures and model specifications. This study also describes the expansion of IDRACS to include a homegrown risk-needs-responsivity assessment.

Authors

• Christopher Inkpen, RTI International

• Nicholas Powell, Georgia Department of Community Supervision

• Bria Cross, Georgia Department of Community Supervision

• Caroline Kery, RTI International

• Pam Lattimore, RTI International