adjusting decision scope and managing missions at
the force level.
Once a tactical military force faces a complex operational problem space, future automated BMAs could
establish a more holistic and wider decision scope and
support resource management at both the platform
and force levels. Ultimately a variety of automated
BMAs could support resource usage at different levels.
BMAs supporting specific sensors and weapons could
be orchestrated by a higher-level BMA architecture.
Thus a system of BMA systems could be implemented.
Resource management is a primary focus of tactical
decision-making and, consequently, a primary application for automated BMAs. The previous section
characterized the battle management problem space
in terms of decision-making; made the distinction
between decisions made by humans and how automated decision aids can support those decisions; and
characterized battle management complexity. This
section looks at some specific concepts for how AI
technologies and concepts can enable and improve
Defining warfare assets (ships, aircraft, submarines,
weapons, sensors, communication devices/networks,
data processing, and jammers) as systems allows
them to be considered as resources and viewed in
terms of their functions, performance, behavior,
structure, and interfaces. It enables quantitative
analyses to be performed based on resource charac-
teristics such as location, status, and expected capa-
bilities. As operations grow in complexity, AI meth-
ods could be used to determine the effective use of
warfare resources when multiple objectives exist that
overlap and conflict. Warfare resource utilization
could, with the aid of BMAs, include forming collab-
orations among systems to enable systems of systems
(or force-level) behaviors and capabilities to better
address complex tactical missions.
Resource management as part of the data fusion
process (Steinberg, Bowman, and White 1998) is
highlighted in figure 11. In this architectural concept, resource management is considered as level 4
processing — assessing the products of data fusion to
determine how to best manage or task resources.
Resource management also provides feedback to the
data fusion process, tasking the level 0– 3 processes.
This data fusion architecture is still a useful paradigm
for implementing AI methods in each processing level. Given significant advances in computing power
and many new sources of data, the use of machine
learning and deep learning can improve resource
management — especially given complex operational situations and distributed warfare resources.
Another way to conceptualize resource management is in terms of systems or data models.
Figure 11. Resource Management within the Data Fusion Architecture.