Using Data to Accelerate Live, Virtual, and Constructive Training in the Marine Corps

Training developers lack methods for determining the benefits of integrating live, virtual, and constructive training. The United States Marine Corps future shift from predominately live training towards the increased use of Live, Virtual, Constructive (LVC) training will require a significant investment. A capital infusion of this nature demands a quantifiable return on investment (ROI) that justifies increasing LVC training costs. Training and education command’s 2020 LVC ROI study identified significant potential readiness benefits and cost savings; however, the data to validate these forecasts lacks fidelity, accuracy, and availability. Part of the issue is that, synthetic training lacks consistent, reusable, and standards-based training materials that are required to maximize the effectiveness of fielded training systems.

Training support packages (TSPs) consist of a stand-alone, exportable package that integrates training products, resources, and materials necessary to plan, prepare, execute, and assess operating force training. Since 2019 MITRE has assisted the Marine Corps in developing a standard TSP that enables faster skill acquisition and improved engagement. This has led to the development of the TSP Rapid Generation Tool (TRGT) that enables the generation of consistent, reusable, standards-based training materials to maximize the effectiveness of fielded training systems while minimizing training preparation time. While TSPs guide what is being trained, measuring the proficiency impacts of the training itself offers a third wrinkle in this problem space.

Training and Education Command is considering the adoption of a scaled performance evaluation measurement system (SPEMS) to be used across tasks and missions to measure training audience proficiency. SPEMS is a 5-point scale that relies on the power of behaviorally anchored ratings scales (BARS) to reliably and accurately appraise task performance. SPEMS leverages the Corps’ thorough training and readiness task construct while adding the layer of numerical granularity that is necessary to reliably measure performance. The results of testing SPEMS demonstrate it provides a more consistently reliable performance evaluation metric than current approaches.

With these findings in place, the Marine Corps needs data scientists, software engineers, and system architects to help build a persistent data strategy to correlate the expenditure of training resources to their readiness impacts.