Context-Driven Information Fusion for Operational Decision Making in Humanitarian Logistics
Context-Driven Information Fusion for Operational Decision Making in Humanitarian Logistics
Alexander Smirnov, Alexey Kashevnik, Tatiana Levashova, Nikolay Shilov
Abstract. The paper describes an approach to information fusion based on the
context management aimed at supporting decision making for humanitarian logistics
operations. Application of GIS as one of the major information sources
is discussed in detail. The paper describes an application of knowledge logistics
as an intelligent information fusion service for creation efficient routing plans
(as one of the major logistics tasks in virtual supply network management) under
given constraints and preferences. The implementation is based on the concept
of open services in a distributed environment of a networked organization.
This application is illustrated via a case study of delivering hospital supplies to
the disaster site.
Keywords: decision making support, ontology, context, context versioning,
humanitarian logistics.
Introduction
User-centric decision support is of high importance for disaster relief &
evacuation operations where multiple participants have to collaborate most efficiently,
and the coalition is often unstable (participants may leave and come
continuously). The quality of decision making depends upon the quality of
information at hand. Problems with information (outdated, incomplete, unreliable,
etc.) form a major constraint in decision making. The practice shows
that one of the most difficult steps in responding to such situations is providing
for the right relief supplies to the people in need at the right time. At the
same time delivering of too much supplies or wrong supplies means losing
time and money [1]. Therefore, humanitarian logistics standing for processes
and systems involved in mobilizing people, resources, skills and knowledge to
help vulnerable people affected by natural disasters and complex emergencies,
is a central issue for disaster relief [2].
Logistics systems play an important role in humanitarian operations.
An intelligent decision support based on the technology of knowledge management may significantly enhance the logistics system capabilities (e.g., reduce costs and time of supplies delivery). Evidently, one of the key information sources for such a system is a geographical information system (GIS) providing
for such information as region geography, available transportation means
(roads, their characteristics and availability), existing infrastructure and facilities,
navigation-related information, etc.
The paper describes an approach to information fusion based on context
management supporting decision making for humanitarian logistics operations.
GIS is used in the approach as one of the major information sources.
Based on the information provided by GIS, creation of efficient routing plans
(as one of the major logistics tasks in disaster relief and evacuation operations)
under given constraints and preferences is performed. The implementation
is based on the concept of open services in a distributed environment of a
networked operation. This application is illustrated via a case study from the
above mentioned domain of disaster relief and evacuation.
The methodology presented proposes integration of environmental information
and domain knowledge in a context of current situation. It is done
through linkage of representation of this knowledge with semantic models of
information sources providing for information about the environment.
The methodology (Fig. 1) considers context as a problem model based on the
knowledge extracted from the application domain and formalized within an
application ontology by a set of constraints. The set of constraints, additionally
to the constraints describing domain knowledge, includes information
about the environment and various preferences (user defined constraints) of
the user concerning the problem solving. Humanitarian logistics as a coalition
operation assumes different user roles. The methodology considers different
user’s roles as different levels of user’s responsibility. The problem is suggested
to be modeled by two types of contexts: abstract and operational. Abstract
context is an ontology-based model integrating the problem relevant information
and knowledge. Operational context is an instantiation of the
abstract context with data provided by the information sources.
The paper is structured as follows. The following sections 2-4 describe the
scenario of the proposed fusion-based decision support; consider all the stages
of the user request processing as well as the context versioning as one of the
core technologies in fusion-based decision support systems (DSS). The reason
for this is that changing environmental conditions should be taken into account
at the moment of decision making. For example, availability of the
transportation routes, possibilities to use certain transport, etc. should be taken
into account by GIS involved into the decision support. Structure of the problem
can also be altered by the changes in the environment. In this case it is
necessary to maintain consistency of the contexts and information sources for
their reuse and definition of missing information and knowledge.
General Scenario of the Fusion-Based Decision Support
The overall scenario of context-driven information fusion in the presented decision
support system (DSS) is shown in Fig. 2. Below, a detailed description
of the scenario is presented.
Before a request can be processed an application ontology describing the
problem the request is related as one to have been built (4). The application
ontology combines the domain knowledge described in domain ontology (3)
and problem solving knowledge described in the task and methods ontology
(2). The domain ontology, in turn, is built based on existing ontologies describing
related domains (1). At these stages fusion of structural knowledge
and problem solving knowledge is performed. At the moment this is proposed
to be done by experts supported by previously developed ontology
management environment [3].
Decision making in dynamic domains is characterized by a necessity to dynamically
process and integrate information from heterogeneous sources and
to provide the user with context-driven assistance for the analysis of the current
situation. Systems of context-driven decision making support are based
on usage of information / data, documents, knowledge and models for problem
identification and solving. This requires an access to electronic documents
complementing the domain knowledge.
Alexander Smirnov, Alexey Kashevnik, Tatiana Levashova, Nikolay Shilov
 |
| Decision Making in Humanitarian Logistics |
Abstract. The paper describes an approach to information fusion based on the
context management aimed at supporting decision making for humanitarian logistics
operations. Application of GIS as one of the major information sources
is discussed in detail. The paper describes an application of knowledge logistics
as an intelligent information fusion service for creation efficient routing plans
(as one of the major logistics tasks in virtual supply network management) under
given constraints and preferences. The implementation is based on the concept
of open services in a distributed environment of a networked organization.
This application is illustrated via a case study of delivering hospital supplies to
the disaster site.
Keywords: decision making support, ontology, context, context versioning,
humanitarian logistics.
Introduction
User-centric decision support is of high importance for disaster relief &
evacuation operations where multiple participants have to collaborate most efficiently,
and the coalition is often unstable (participants may leave and come
continuously). The quality of decision making depends upon the quality of
information at hand. Problems with information (outdated, incomplete, unreliable,
etc.) form a major constraint in decision making. The practice shows
that one of the most difficult steps in responding to such situations is providing
for the right relief supplies to the people in need at the right time. At the
same time delivering of too much supplies or wrong supplies means losing
time and money [1]. Therefore, humanitarian logistics standing for processes
and systems involved in mobilizing people, resources, skills and knowledge to
help vulnerable people affected by natural disasters and complex emergencies,
is a central issue for disaster relief [2].
Logistics systems play an important role in humanitarian operations.
An intelligent decision support based on the technology of knowledge management may significantly enhance the logistics system capabilities (e.g., reduce costs and time of supplies delivery). Evidently, one of the key information sources for such a system is a geographical information system (GIS) providing
for such information as region geography, available transportation means
(roads, their characteristics and availability), existing infrastructure and facilities,
navigation-related information, etc.
The paper describes an approach to information fusion based on context
management supporting decision making for humanitarian logistics operations.
GIS is used in the approach as one of the major information sources.
Based on the information provided by GIS, creation of efficient routing plans
(as one of the major logistics tasks in disaster relief and evacuation operations)
under given constraints and preferences is performed. The implementation
is based on the concept of open services in a distributed environment of a
networked operation. This application is illustrated via a case study from the
above mentioned domain of disaster relief and evacuation.
The methodology presented proposes integration of environmental information
and domain knowledge in a context of current situation. It is done
through linkage of representation of this knowledge with semantic models of
information sources providing for information about the environment.
The methodology (Fig. 1) considers context as a problem model based on the
knowledge extracted from the application domain and formalized within an
application ontology by a set of constraints. The set of constraints, additionally
to the constraints describing domain knowledge, includes information
about the environment and various preferences (user defined constraints) of
the user concerning the problem solving. Humanitarian logistics as a coalition
operation assumes different user roles. The methodology considers different
user’s roles as different levels of user’s responsibility. The problem is suggested
to be modeled by two types of contexts: abstract and operational. Abstract
context is an ontology-based model integrating the problem relevant information
and knowledge. Operational context is an instantiation of the
abstract context with data provided by the information sources.
The paper is structured as follows. The following sections 2-4 describe the
scenario of the proposed fusion-based decision support; consider all the stages
of the user request processing as well as the context versioning as one of the
core technologies in fusion-based decision support systems (DSS). The reason
for this is that changing environmental conditions should be taken into account
at the moment of decision making. For example, availability of the
transportation routes, possibilities to use certain transport, etc. should be taken
into account by GIS involved into the decision support. Structure of the problem
can also be altered by the changes in the environment. In this case it is
necessary to maintain consistency of the contexts and information sources for
their reuse and definition of missing information and knowledge.
General Scenario of the Fusion-Based Decision Support
The overall scenario of context-driven information fusion in the presented decision
support system (DSS) is shown in Fig. 2. Below, a detailed description
of the scenario is presented.
Before a request can be processed an application ontology describing the
problem the request is related as one to have been built (4). The application
ontology combines the domain knowledge described in domain ontology (3)
and problem solving knowledge described in the task and methods ontology
(2). The domain ontology, in turn, is built based on existing ontologies describing
related domains (1). At these stages fusion of structural knowledge
and problem solving knowledge is performed. At the moment this is proposed
to be done by experts supported by previously developed ontology
management environment [3].
Decision making in dynamic domains is characterized by a necessity to dynamically
process and integrate information from heterogeneous sources and
to provide the user with context-driven assistance for the analysis of the current
situation. Systems of context-driven decision making support are based
on usage of information / data, documents, knowledge and models for problem
identification and solving. This requires an access to electronic documents
complementing the domain knowledge.
0 Response to "Context-Driven Information Fusion for Operational Decision Making in Humanitarian Logistics"
Post a Comment