Main Subcommittee Category: Policy, Standards, Management, and Acquisition

How would you label your submission?: Government

Abstract:

Data-driven decision-making and big data have become ubiquitous in Department of Defense (DoD) as there is widespread acknowledgement of the potential for both to advance warfighter performance (DoD Data Strategy, 2020). Advances in artificial intelligence (AI), machine learning (ML), and cloud-based data storage and processing have created opportunities to conduct previously impossible analyses. Advances in wearable and non-invasive physiological sensors, along with eye tracking technology, can be utilized with AI/ML to objectively capture important human performance measurement (HPM) that currently require human observation. While these capabilities hold significant promise for advancing DoD data analytic tools, substantial groundwork must be performed to ensure the reliability and validity of the underlying data. Fortunately, the DoD has made significant strides in this space by investing in data standards (e.g., Experience Application Programming Interface, Human Performance Modeling Language) that delineate the types and format of system-based data required to better understand warfighter learning and performance (Poeppelman et al, 2013). This paper is intended to revisit the progress made on learning and HPM standards as an essential capability for data strategy and reframe their function as part of the larger DoD wide data strategy guiding principles: ‘collective data stewardship’ and ‘data collection’. The ultimate goal is to ensure learning and HPM data standards include language to capture and utilize reliable, valid, and transparent data from the beginning of training and throughout the performance of their duties. Now is also an opportune time to accommodate emerging technologies and the unique potential they offer to close data gaps, and meaningful visualizations for a variety of stakeholders. By applying Poeppelman and colleagues’ (2013) model to a beginning stage naval aviator training use case, the authors will conceptualize extensions for advanced learners and provide initial recommendations for changes to the standards that account for emerging technologies.

Keywords: AI;ANALYTICS;ASSESSMENT;BIG DATA;HUMAN PERFORMANCE;STANDARDS

Sarah Beadle

Affiliation

Refine Results

Automated Human Performance Measurement: Standardizing Lifelong Learning Training Data