Improving Soldier Factors in Prediction Models

Period of Performance: 12/17/1998 - 06/16/1999

$120K

Phase 1 SBIR

Recipient Firm

Micro Analysis and Design, Inc.
4949 Pearl East Circle, Suite 300
Boulder, CO 80301
Principal Investigator

Research Topics

Abstract

Increasingly, resource allocation questions are being answered with computer modeling and simulation. Additionally, much of the training provided to soldiers relies on computer models and simulations of the combat environment. To date, these models do not adequately address the effects of training on performance, nor do they adequately address the interactions between factors that shape human performance. This effort will remedy that situation by developing methods and data structures that will allow computer generated force models (CGFs) to adequately respond to the key features of training. We propose a mix of model development and data collection. The models we propose reflect the state-of-the-art in human performance modeling and what we know about the effects of training on performance. These models are based on research and our extensive experience in these areas. The data sources we propose represent a get something in place now and build on it later approach where we start with subject matter expert data in Phase I and then expand our data search in Phase II. As such, this proposal presents an innovative approach to ensuring that improved CGFs are in place soon, while also providing room for planned growth as we learn more through future research. BENEFITS: The Immediate benefits to the military are 1) that models will be able to to be used to evaluate the payoff of a training investment in terms of resulting soldier performance and 2) the models used for training soldiers will be more realistic. For commercial applications, the model structures proposed are generalizable and could allow industry to make trades between the value of training vs. other strategies for process improvement such as automation.