With the increasing availability computer-based simulations, military Courses of Action (COA) planning can go beyond the doctrinally required three distinct COAs to accomplish the mission. In principle, planners can generate, simulate, compare and select from a large number of COAs, thus vastly increasing the chances for finding good COAs. However, any planning based on simulations has to face an intrinsic problem of simulation models, namely, that there is an inevitable gap between the models and reality. Comparing and selecting COAs on the basis of simulation results can be problematic in the face of simulation model gaps and errors. The gap between model and reality is not an issue for computer simulations alone; as Secretary Rumsfeld’s now-famous remark about “unknown unknowns” suggests, simulations that human minds perform are also subject to the same problem. However, humans have evolved a set of heuristics by which they sometimes critique their models and simulations to test their robustness. The awareness of this gap is not common in research on decision support for planning involving computer simulation. In addition, since much of the simulation takes place rapidly in the computer, the planner is not even aware of possible gaps, and there is a tendency to place too much reliance on the results of the simulation, and hence on the plans based on them. A related issue is in how to handle the uncertainties in the models used in simulation. A standard approach has been to assume probability distributions and make decisions based on expected values for outcomes of interest. In fact what the planner needs is not quite the expected values, but rather what effect the uncertainties have on what dimensions of the outcome to what extent, and correspondingly, how to make the desired outcomes be less sensitive to the uncertainties. Both of these issues – handling the gaps as well the uncertainties – require a shift in point of view from optimality to robustness. In order to realize the full potential of the vastly increased search spaces made possible by computer simulation, it is essential that the decision support system empower the planner to explore the plans for robustness of the selected plans. The research challenge is to identify, and incorporate as part of decision support systems, a variety of techniques by which the selected plan can be tested for sensitivity to
[1]
Cheng Blanco,et al.
Social Science Computer Review
,
1987
.
[2]
Giorgio Rizzoni,et al.
An Architecture for Exploring Large Design Spaces
,
1998,
AAAI/IAAI.
[3]
Kaisa Miettinen,et al.
Nonlinear multiobjective optimization
,
1998,
International series in operations research and management science.
[4]
Balakrishnan Chandrasekaran,et al.
A family of dominance filters for multiple criteria decision making: choosing the right filter for a decision situation
,
2001
.
[5]
Steven C Bankes,et al.
Tools and techniques for developing policies for complex and uncertain systems
,
2002,
Proceedings of the National Academy of Sciences of the United States of America.
[6]
Paul K. Davis,et al.
Effects-Based Operations (EBO): A Grand Challenge for the Analytical Community
,
2006
.
[7]
Eric Heilman,et al.
FROM SIMULATION TO INSIGHTS: EXPERIMENTS IN THE USE OF A MULTI-CRITERIAL VIEWER TO DEVELOP UNDERSTANDING OF THE COA SPACE
,
2003
.
[8]
Eric Heilman,et al.
Mining simulation data for insights about a decision space: application to an urban combat COA
,
2004,
SPIE Defense + Commercial Sensing.