European Journal of Operational Research 55 (1991) 403-408 403

North-Holland

Decision Support Systems for military

applications
Fatima Cristina Canazaro Dargam
Instituto Militar de Engenharia (IME), P~a. General Tibfircio 80, Praia Vermelha, CEP 22290 Rio de
Janeiro, Brazil, and
Instituto de Pesquisas da Marinha (IPqM), Rua Ipiru 2, Ilha do Governador, CEP 21931 Rio de Janeiro,
Brazil

Emmanuel P. Lopes Passos

Instituto Militar de Engenharia (IME), P~a. General Tibfircio 80, Praia Vermelha, CEP 22290 Rio de
Janeiro, Brazil

Fernando Da Rocha Pantoja

Instituto de Pesquisas da Marinha (IPqM), Rua lpiru 2, Ilha do Governador, CEP 21931 Rio de Janeiro,
Brazil

Received July 1990

Abstract: This paper presents and discusses some fundamental aspects of the implementation of an
Expert Support System (ESS) in the military field of naval electronic warfare. The importance of
knowledge-based systems for decision support applications in the military area is emphasized. A brief
view of the characteristics of military systems, concerning the interaction of information and real systems,
is presented.

Keywords: Decision Support Systems, Expert Support Systems, Artificial Intelligence, military systems,
electronic warfare

I. Introduction effective decision making by the commander (Mc-

Mahon, 1986).
The 'decision making' topic has been widely The objective of this paper is to present a case
discussed in relation to a number of factors. The study of a DSS applied to the specific military
most important one concerns the cognitive con- field of Electronic Warfare (EW) for a naval
straints of men (Bosman, 1987). A person is not environment. In order to describe the role of the
able to properly recognize or define a problem at system to be presented in this paper, some con-
once without the aid of several instruments which siderations about military systems and their char-
transfer data into information. This also happens acteristics must be made.
in military applications. The success in a military A military conflict can be characterized as a
conflict depends on a combination of factors such dynamic system composed of an information sys-
as power, strategy, equipment and most of all tem and a real system, where the former controls

0377-2217/91/$03.50 © 1991 - Elsevier Science Publishers B.V. All rights reserved

404 F.C.C. Dargam et aL / DSS for military applications

Artificial Intelligence systems have been shown

~I REAL potentially adequate to increase our ability to
--I SYSTEM respond more quickly and efficiently to those new
challenges and threats by performing lower-level
data fusion and reduction tasks on our behalf,
thus enabling us to concentrate more on critical,
ACTION higher-level decision making activities.

DATA
t 2. DSS in the military context

DECISIONl In military systems, support of decision making

requires more than storage and retrieval of iso-
lated facts. It requires the generation of reason-
l INFORMATIONI
- ably inferred propositions which reflect the local
SYSTEM.,]~ conflict scenario as well as the personalities of
the commanders (Dockery, 1982).
A conflict situation is always very dynamic,
Figure 1. A dynamic system
demanding quick and adequate answers within a
wide range of possible options. In this case, the
use of a knowledge-based system, envisaging au-
tonomous decision-making procedures and taking
the latter. The control process is performed
into consideration previous storage experiences,
through data and information transfer. The real
is justified. Consequently, a higher confidence
system sends messages concerning its state to the
factor is expected from the system action. Such a
information system. The information system acts
system can be structured based upon a data base,
according to the obtained information, sending
a model base and a knowledge-based system. It
messages to the real system, as shown in Figure 1,
can also be considered as a combination of a
causing changes in its state by means of actions
conventional Decision Support System and an
derived from decision making processes (Bots,
Expert System (Hayes-Roth et al., 1983), that is,
1987).
an 'Expert Support System' (ESS)(Passos, 1988).
An Information System is an integrated user-
The ESS configuration shown in Figure 2 is
machine system for providing information to sup-
composed of an Information Module, which con-
port operations management and decision mak-
tains a knowledge base, a data base and a model
ing (Bosman, 1987). We can consider a DSS as a
base, and is responsible for supplying the system
special kind of Information System. Its main ob-
with all the formally represented information
jective is the efficiency of the decision making
concerning the specific application; an Inference
within a context, using a computer system as
Engine, which is responsible for the problem
assistance tool.
solution process and reflects the decisions sup-
Among the military applications, Electronic
ported by the system, an Explanation Module,
Warfare is a dynamic and wide area and has had
which explains to the user the system's actions;
a quick growth due to the development of mod-
and a User Interface Module which provides the
ern weapons that are commanded, directed and
user interaction with the system.
controlled electronically (Schleher, 1986). As
these modern weapon systems become increas-
ingly lethal in speed in accuracy, there is a critical
need to supply decision makers as early as possi- 3. Basic concepts of Electronic Warfare
ble with precise and relevant information ex-
tracted from the massive amount of data gener- Electronic Warfare (EW) is the military area
ated, distorted, modified and segmented by each which is responsible for recognizing and weaken-
side involved in military conflicts. ing hostile electromagnetic emissions, making the
 F.C.C. Dargam et al. / DSS for military applications 405

INFORMATION MODULE Counter-CounterMeasures. Figure 3 illustrates

!i
this classification.
KNOWLEDGE DATA MODEL The SIGINT activity searches, receives, lo-
BASE BASE BASE cates, analyzes and identifies electromagnetic
emissions of communication, detection, naviga-
tion, command and control systems of the enemy,
with the objective of obtaining information about
the current situation, the threat and the opera-
tional methods of the enemy. The SIGINT task is
EXPLANATION INFERENCE implemented by two subtasks: COMINT - Com-
MODULE ENGINE munications Intelligence, which refers to commu-
nication interception and interpretation; and
ELINT - Electronics Intelligence, which is re-
sponsible for intercepting and interpreting radar
USER transmissions, control, detection, navigation and
INTERFACE identification systems.
The ESM activity is responsible for the analy-
sis of the electromagnetic environment, for
r-'
I
-i
I
searching for hostile emissions and for the techni-
I USER
I
i
I
cal information needed for the application of the
L----- J adequate countermeasures and counter-counter-
Figure 2. ESS - Expert Support System measures.
The ECM activity is responsible for perform-
ing the selected electronic countermeasure, with
the objective of disturbing the communication-,
electromagnetic actions adopted by the enemy in radar information- and control information sys-
a conflict situation as ineffective as possible. tems of the enemy, making them ineffective and
EW equipments are indispensable components confused.
of a modern weapon system, which applies elec- The ECCM activity protects the system against
tromagnetic energy in different ways (e.g. in sig- ECM activities and, therefore, is responsible for
nal transmission using radio, radar, radio naviga- guaranteeing the integrity of the informations
tion and radio control). processed.
Radar intercepts targets in the air, at sea or in
the battlefield, determining their coordinates with
a high degree of accuracy, and it is widely used
for fire control, guidance of aircrafts and missiles
as well as navigation. Systems such as radar,
E.W.
however, have two major vulnerabilities, which
are explored by the EW analysts: the emissions S I G I NT
can be intercepted and the reception channels
can also capture non-desirable signals, such as COMIN[ ELINT
man-made interferences and false data created
by electronic countermeasure systems.

I ECCM I ECM
3.1. Electronic Warfare classification

EW is generally divided into four major activi-

ties: SIGINT - Signal Intelligence; ESM - Elec- ESM
tronic Support Measures; ECM - Electronic
CounterMeasures and ECCM - Electronic Figure 3. Operational classification of Electronic Warfare
406 E C.C. Dargam et aL / DSS for military applications

4. An ESS applied to a naval E W / r a d a r system

/ ~ ELECTROMAGNETIC
The system proposed in this paper is an ESS EMISSIONS
for deployment in the military area of Electronic
Warfare/Radar. It has the objective of helping
in the decision-making process of an adequate
countermeasure to be performed by the ECM
ESM /Tableof threats
system, aiming at deceiving or neutralizing a SYSTEMSL
threat identified by the ESM system (Dargam,

:Lro CHOOSE ONE

. . . . . . . . -7

1989). It is functionally placed between the EW

divisions of ESM and ECM, as shown in Figure 4. PRIOR
THREAT ,Operatori
i i

Its main objective is to help the EW analyst in the

decision making of a quick and adequate counter- :threatldata ] 1Threat L . . . . . . J

measure in order to neutralize or deceive an

identified threat.
In the context of Electronic Warfare, one of
the major points is the proper representation of
the ignorance of the existing facts (lack of knowl- ACTION
edge). The Expert Support System implemented
recognizes gaps in the actual overall scenario and
searches for the proper minimum necessary piece Figure 5. A conventional Electronic Warfare/Radar system
of information, through interaction with intelli-
gence elements (EW analyst or knowledge base).
It uses this new information or inferred data to In order to understand the need for and the
update its situational model and to properly de- advantages of an ESS applied to an EW system, it
cide about countermeasure actions, taking into is recommended that we analyse first a conven-
consideration the current task or mission, possi- tional E W / R a d a r system, as shown in Figure 5.
ble threats and the conflict situation. In a conventional EW system, when electro-
magnetic emissions are detected, the Electronic
Support Measures (ESM) system identifies the
received signals, comparing their characteristics
[~~ ELECTROMAGNETIC
EMISSIONS with those of other emitters already stored in the
system library of threats (e.g. according to
NATO). As a result of this operation, the ESM
ESM system generates a table with all the identified
threats, their priority levels, their major parame-
SYSTEM ters and their confidence levels (concerning iden-
tification). Then a threat is chosen by the EW
analyst and the ESM system delivers to the Elec-
tronic CounterMeasures (ECM) system all the
relevant data about the selected threat and also a
suggested electronic countermeasure.
SENSORS --I COMPUTER By now considering an Electronic Warfare/

NFORMAT'ONS
I Radar system embodying in its design an Expert
Support System, as shown in Figure 4 and as
functionally described in Figure 6, important de-
cisions about countermeasure actions can be en-
hanced. When an electromagnetic emission is de-
tected, the ESM system identifies the type of
ACTION emission, classifying the threat. The ESS then
Figure 4. Electronic Warfare/Radar system + ESS decides whether the countermeasure suggested
 F.C.C. Dargam et al. / DSS for military applications 407
41-/_ ELECTROMAGNETIC factor, the most adequate countermeasure to the
EMISSIONS identified threat and to the current conflict con-
ditions. The resulting countermeasure action data
ESM
receives /
~ Library
of
is sent to the ECM system to be executed. At this
identifies generic time the ESS also activates a temporal dynamic
threats threats interface with the Naval Electronic Warfare
© equipment, allowing a continuous update of the
countermeasure techniques applied. This is done
TABLE L 'r~ .......
E.W. by testing the effectiveness of the countermea-
OF j- 'i sure and the continuing existence of the identi-
THREATS ',OPERATOR,,
i i
fied threat by means of sensors, and works as a
L . . . . . . . i

1 threat is chosen positive feedback to the whole system (Dargam,

1989).
provides the According to the given functional description
threat data
+ a suggested of the EW system, the ESS proposed has its
countermeasure

r . . . . . -i
i i
i USER l
NO Library I I
of
specific
threats
© USER ]
I ESS INTERFACEI
Refines the
countermeas-
ure actions
INFORMATION DATA
MODULE MEMORY
-t
Executes theI
counterm,eas- l
TION J
MODULE ]
ure actions ]

effective
ESS
Checks the
efficiency of
the counter-
measure action
threats not
present anymore Q
__•N' FERENCE I
ENGINE I

not effective COMMAND.c

DYNAMIC
AND
© INTERFACE
-7 WITH THE
MESSAGES

Figure 6.Asimplified functional system description i --~1 TO/FROM

THE E.W.
t E.W. --
SYSTEM SYSTEM
by the ESM system should be accepted or not,
considering the general characteristics of the
threat and the current conflict situation. If the
result of this decision process is the refinement of ELECTRONIC
the countermeasure, then the ESS performs this WARFARE/RADAR
SYSTEM
refinement considering the threat's specific pa- L . . . . . . . . . .

rameters in order to infer, with a high certainty Figure 7. ESS applied to Electronic Warfare
408 F.C.C. Dargam et al. / DSS for military applications

original configuration (see Figure 2) adapted to References

the needs of the application, as shown in Figure Bosman, A. (1987), "Relations between specific Decision Sup-
7. port Systems", Decision Support Systems 3/3, 213-224.
A simulation of this Expert Support System Bots, P.W.G., and Sol, H.G. (1987), "An environment to
was implemented (Dargam, 1989) in OPS5 (Offi- support problem solving", Decision Support Systems 3/3,
cial Production System - version 5) (Forgy, 1981), 225-231.
Ciganer, P.A. (1986), "AI technology expectations and reality",
for a microcomputer environment (Velasco, 1987). Journal of Electronic Defense 8.
Due to the obviously highly confidential contents Coelho, H. (1988), "State of the art of DSS's in 1988",
of the ESS components, details of this implemen- Proceedings of the First International Informatics Congress,
tation are not discussed in this paper. Rio de Janeiro, Brasil, 113-115.
Dargam, F.C.C., Passos, E.P.L., and Pantoja, F.R. (1988),
"Desenvolvimento do sistema SEGE - - Sistema Especial-
ista aplicado ~ Guerra Eletr6nica', Proceedings of the 5th
5. Conclusions Artificial Intelligence Brazilian Symposium, Natal, RN.
Dargam, F.C.C. (1989), "Sistema especialista aplicado
Artificial Intelligence plays an important role Guerra Eletrfnica", Masters dissertation, IME-RJ, Brazil.
providing new concepts for the design of Decision Dockery, J.T. (1982), "Fuzzy design of military information
systems", International Journal of Man - Machine Studies 6,
Support Systems, as for example the Expert Sup- 1-38.
port Systems (ESS). The experience being ob- Forgy, C.L. (1981), "OPS5 user's manual", Technical Report
tained in the first attempts to build ESS will CMU-CS-79-132, Carnegie-Mellon University, Pittsburgh,
certainly provide DSS researchers with worth- PA.
while results in terms of effectiveness. Hafner, A.N., and Sutherlin, C. (1986), "EW decision support
technology", Journal of Electronic Defense 10, 91-96.
The importance of the use of Decision Sup- Hayes-Roth, F., Waterman, D.A., and Lenat, D. (1983),
port Systems based on Expert Systems for mili- Building Expert Systems, Addison-Wesley, Reading, MA.
tary applications was characterized. McMahon, T.L. (1986), "Decision making in modern war",
A case study in the Electronic Warfare area Military Review 10, 33-41.
was presented, describing a dedicated configura- Passos, E.P.L. (1988), "The Artificial Inteligence contribution
to Decision Support Systems", Proceedings of the First
tion of an Expert Support System for this applica- International Informatics Congress, Rio de Janeiro, Brasil,
tion. 109-112.
It has become clear that Decision Support Schleher, D.C. (1986), Introduction to Electronic Warfare,
Systems are an emerging military technology Artech-House.
Velasco, F.R.D. (1987), "Uma implementa~o da linguagem
which will provide the necessary information han-
OPS5 para computadores compatlveis com IBM-PC", Pro-
dling and analysis tools in a very near future. ceedings of the 4th Artificial Intelligence Brazilian Sympo-
sium, Uberl~ndia, MG, Brazil.