Predictive Modeling in the Military: Similar Goals, Divergent Paths

Summary

This project reported here had the objective of answering four specific questions:

• Do predictive models that have been created for military installations work?

• Can they be refined to work better?

• Are they sufficiently accurate so that land managers and State Historic Preservation

officers can use them in evaluating management decisions about installation resources?

• Can a predictive model be integrated into a more dynamic operational model that would be useful across the DoD to increase cost efficiency of cultural resource management at large installations?

The proposal defined a three-step process for answering these questions. The first step was to determine the pervasiveness of predictive modeling in the military through a questionnaire sent to installations representing all branches of the service. Although not intended to be a complete canvassing of the military use of predictive models, the objective of the questionnaire was to achieve a reasonable sample from which inferences could be drawn. The second step was to choose models from four of the responding installations for an in-depth evaluation of their technical quality, accuracy, and general utility as a management tool. This evaluation was to include suggestions about ways to improve the utility of each model. The proposed third step was take the results of the evaluation and design a follow-up study that could directly contribute to more effective use of predictive modeling by military installations.

Below we describe the questionnaire and the selection process. The next section of this report provides an evaluation of each of the four selected predictive models. In the final section of the report, we suggest what we would propose as a useful follow-up study.

This Resource is Part of the Following Collections

• SRI Foundation

Keywords

Individual & Institutional Roles

Contact(s): Jeffrey Altschul

File Information