Content uploaded by Adam Widera
Author content
All content in this area was uploaded by Adam Widera on May 24, 2017
Content may be subject to copyright.
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
Integrated Logistics and Transport
Planning in Disaster Relief Operations
Adam Widera
Chair of Information Systems and Supply
Chain Management,
University of Münster
adam.widera@ercis.uni-muenster.de
Sandra Lechtenberg
Chair of Information Systems and Supply
Chain Management,
University of Münster
sandralechtenberg@uni-muenster.de
Gaby Gurczik
German Aerospace Center
Institute of Transportation Systems
gaby.gurczik@dlr.de
Sandra Bähr
German Aerospace Center
Institute of Transportation Systems
sandra.baehr@dlr.de
Bernd Hellingrath
Chair of Information Systems and Supply Chain Management,
University of Münster
bernd.hellingrath@ercis.uni-muenster.de
ABSTRACT
Decision making in the area of humanitarian logistics and supply chain management often suffers because of the
interrelations between planning horizons, tasks, and crisis management lifecycle phases. In this paper, we
present a method, an exemplary prototypical implementation and its evaluation within a relief organization.
Based on a structured literature analysis (a review of existing information systems as well as a consideration of
ongoing research projects), basic requirements for an integrated logistics and transport planning approach were
derived. Together with end-user involvement, these results were used to design and prototype a concept of an
appropriate information system, which was applied and evaluated in a tabletop exercise. The generated results
are promising in terms of having a positive impact on the logistics effectiveness. In combination with the
identified limitations, our results promise to have an impact on future ISCRAM research.
Keywords
Humanitarian Logistics, Logistics Planning, Transport Planning, Use Case, Simulation, Routing
INTRODUCTION
Natural disasters affect significantly human kind. In the last years, Europe was troubled by several natural
disasters (e.g. extreme flooding in June 2013 in Central Europe), which become more frequent and cause more
damage (Bevere et al., 2013). The emerging nature of disasters has a deep impact on the requirements for
humanitarian operations.
Being the main driver for the effectiveness and efficiency, the humanitarian logistics preparedness level of all
involved actors, from non-governmental organizations (NGOs), over governmental organizations up to
commercial providers, plays an important role for the responsiveness to major disasters. Procurement activities
can sum up to 60 % of the total sum spend by humanitarian organizations and backlogs can lead even to break
downs of operations. Another success factor of humanitarian operations is the ability to source goods from pre -
positioned inventories. It is not only the most important time advantage that can be utilized here, but –as pre-
positioning networks examples like the UNHCR show –well planned inventories offer a range of response
coordination and cost benefits. As the practices of DG ECHO show, the importance of transportation support is
752
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
considered as a crucial function of humanitarian operations. The transportation system is of outstanding
importance not only for the mobility and the supply of the population, but also for professional responders who
depend on functioning and reliable transportation infrastructures to reach the corresponding action places, to
ensure evacuation and to provide the affected population as well as the emergency forces with goods and
services.
Agility is a key attribute for logistics and transport planning. Information Systems can improve the preparedness
and the responsiveness of humanitarian operations in Crisis Management by contributing to more agile
responder coordination. As preconditions, information systems must have the abilities 1) to diagnose the
divergence between the actual and the expected situation (detection) and 2) to adapt the current strategy
coherent with the diagnostics (adaption). Moreover, it is vital that both tasks have to be performed fast enough
and in a pertinent and accurate way to ensure a timely and effective reaction to the situation (reactivity and
effectiveness).
The aim of the paper is to explore, how the combination of information systems from different phases of the
crisis management cycle, functions and planning horizons can lead to a more agile crisis response and
management. Therefore, the DLR and the WWU initiated a cooperation for the exploration of joined logistics
and traffic management information systems within the EU FP7 project DRIVER (2014)*. The systems provide
relevant information for crisis managers to cope with challenges of the logistics and transport chain during the
preparation, respectively response phase in the crisis management cycle.
Within the paper, we first provide the results of a structured literature review and give then an extensive
overview on available information systems for the planning (preparation) and response phase as well as on
related work, concerning integrated logistics planning and transportation management systems. The goal is, to
carve out the potential of integrating such systems, including identifying chances and risks, as well as to depict
requirements and framework conditions for the implementation. In the second part of the paper, we present a
concept of how such a connected system could look like. The theoretical consideration is complemented by a
prototypical implementation and its evaluation of operating experience collected during a tabletop exercise
performed with the Federal Agency for Technical Relief (THW).
RELATED WORK
Research in crisis management and humanitarian logistics has increased over the years, especially since the
tsunami catastrophe in Southeast Asia in 2004. A growing part of this research is dedicated to the usage of
information and communication technology in the field (Altay and Green, 2006). Humanitarian organizations
more and more recognize the fact that also humanitarian supply chains can profit from and should be supported
by supply chain management systems. In the year 2004, the International Committee of the Red Cross (ICRC)
was the first organization starting to implement an IT-system comprehensively supporting them in tasks
connected to the material flow in the supply chain. Nonetheless, only every fourth organization uses such a
system while the others rely on spreadsheets or even pencil and paper (Blecken, 2010).
To get an overview of the current research on decision support systems in the context of disaster relief logistics,
a structured literature search has been conducted. Three requirements, which a source needs to fulfill to be
counted as relevant, have been defined:
1. It has to describe a tool, model or idea, which aims at supporting decision makers, i.e. provide
decision support in some way.
2. The decisions to be made, resp. the problem dealt with, should specifically be within the scope of
logistics or supply chain management. Sources dealing with general crisis management problems
will not be considered relevant.
3. Finally, the environment in which these decisions need to be made should exhibit features of a
crisis or disaster situation.
Accordingly, the following search term has been defined:
“decision support” AND (crisis OR disaster OR emergency OR “relief operation” OR
humanitarian) AND (logistics OR “supply chain”)
Since all sources only presenting a model, which can be applied as a mean of decision support, or suggesting
ways to enhance sense making in a crisis situation should be included in the literature review, two further search
terms have been built by replacing “decision support” with “decision model” as well as “sense making”. The
resulting three search terms have been applied to three literature databases: EBSCOhost, ScienceDirect and
Scopus. The search has been limited to tiles, keywords and abstracts. Apart from this, no further restrictions
(e.g. specific years of publication) have been selected. After a manual review of the results’ abstracts, relevant
753
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
sources have been identified, which were used to provide an overview on how the topic of decision support in
crisis logistics is treated in research.
Decision support systems developed and described in research mostly deal with problems from the preparedness
and response phase. The most prominent problems handled are the facility location problem (e.g. Charles et al.,
2016; Degener et al., 2013) and the delivery scheduling problem (e.g. Ben Othmann et al., 2014; Kuo et al.,
2015). Very little research has been done on systems supporting decisions in the mitigation and recovery phases.
An exception are Fallucchi et al. (2016) and Lorca et al. (2016), who deal with debris-related problems.
Moreover, the results of the literature search show that decision support systems developed for logistics in crisis
management in research are usually addressing one specific problem and therefore one management level and
one crisis phase. Although it has already been recognized that decisions at different management levels or cycle
phases are dependent on the ones taken at other levels resp. in other phases (e.g. Irohara et al., 2013; Kelle et al.,
2014), only a very limited number of sources suggest decision support systems or models producing integrated
recommendations across the crisis management cycle (Ransikarbun and Mason, 2016). Exceptions to this
observation are for example, the stochastic programming model by Garrido et al. (2015), which aims at
optimizing the inventory level of prepositioned relief items and the flow of those items after a flood crisis and
consequently integrates decision problems from the preparedness and response phase.
Regarding the methods which are applied to provide support to decision makers, the by far most used one is
mathematical optimization, including goal programming, stochastic programming etc. Especially when
specifically considering sources which propose or even implement a system –in contrast to describing a model
or framework –the minority applies techniques from other fields and an even smaller fraction of them use
simulation. While most of those sources use simulation (e.g. D’Uffizi et al., 2015; Fikar et al., 2016b), there are
others which combine simulation with optimization (Ben Othmann et al., 2017; Fikar et al., 2016a). Problems
treated in these sources are all connected to the response phase and mostly deal with delivering relief items to
people in need or more general with the status of the road network, i.e. with the question which effect sudden
damages of single elements will have on the whole network. Fikar et al. (2016a) also investigate a collaboration
between private and relief organizations.
Overall it can be observed, that decision support systems proposed in literature typically apply optimization
techniques to solve a specific and restricted problem. Even though there are exceptions, still most sources follow
this pattern. The usage of exact solution procedures deprives the possibility to include decision maker’s
experience and observations into the decision-making process. Moreover, optimization models are often not
sufficient to depict the high complexity and uncertainty of a crisis situation. Especially the formulation of an
adequate objective function is difficult because in a crisis different factors which are hard to express
quantitatively, such as the life or suffering of people, play a major role in decision making. Simulation might be
a technique to overcome the shortcomings of solely applying optimization (Ben Othmann et al., 2017; D’Uffizi
et al., 2015). As a mean for supporting decision-making, simulation has already been widely researched in the
context of commercial logistics (compare e.g. Terzi and Cavalieri, 2004). Reasons for utilizing simulations as a
decision support tool are e.g. that a simulation model facilitates understanding of the real system and its
behavior, provides a basis for discussion about it and is able to show previously hidden relationships and
interdependencies. Moreover, the main advantage is the possibility to perform “what-if” analyses and be able to
estimate a decision’s effects before actually implementing it (Semini and Fauske, 2006). Nonetheless, for
simulated solution proposals to be relevant, planning tools need to incorporate information from operative tools,
which often provide exact solutions. Such a combination of simulation and exact solution techniques or
operative tools can lead to the capability of adequately handling the complexity of a crisis situation (Fikar et al.,
2016a).
Apart from decision support systems proposed in literature, there are also research projects, existing products
and experiments dealing with the topic of decision support in disaster relief logistics. In the context of such
projects, various tools are developed and suggested to serve as a mean of decision support for crisis logistics
managers. The following table gives an overview about the tools obtained through an internet search and its
intended crisis management lifecycle. Many tools for crisis management and decision support exist but with
different levels of maturity. Due to a limited accessibility and varying available information, a rather general
overview of the identified tools is provided. The tools identified in the search vary in terms of scope, purpose
and degree of actual deployment. Further, it has to be noted that the list does not claim to be exhaustive, but it is
meant to provide a first overview of the current landscape on logistics management and transportation tools.
754
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
Tool
Short description
Preparation
phase
Response
phase
Recovery
phase
ELITE
web-based ‘living document’ for post-
crisis lessons learning
X
CRISMA
simulation-based decision support
system
X
FORTRESS
incident evolution tool that will assist
in forecasting potential cascading
effects
X
CRIMSON
generic platform to train crisis
managers and field crews
X
SECTOR
secure European common information
space (CIS)
XX
KOKOS
IT tools to involve the active
participation of the general public for
self-help activities
X
Disaster LAN
Web-based task, mission and resource
management system (only in the U.S.)
X
SAHANA
Web-based collaboration tool for
disaster response coordination between
governments, NGOs etc.
XX
Fusionpoint
a web-based collaboration tool between
emergency operation centers
X
EmerGeo
(EmerGeo
Maping)
Common Operating Picture helping
government and industry to mitigate,
prepare for, respond to and recover
XXX
RIB / LUPP
tool for decision-making and follow-up
of emergency response operations
X
Ushahidi
Map-based web application to
crowdsource data from people on the
ground to aid relief efforts
XX
SensePlace2
map-based web application that
integrates multiple text sources
(Tweets, News, Bloqs etc.)
XX
Table 1. Overview of decision support tools for crisis management (Boin and Schaap, 2010; Chan, 2013;
CRISMA, 2015; ELITE, 2014; FORTRESS, 2014; KOKOS, 2015; Morrow and Bachoura, 2013;
SECTOR, 2014)
In the project CRIMSON a system was developed that offers an effective and generic platform combining
simulation and virtual reality technologies for inter-organizational preparation, rehearsal and management of
crisis situations, in order to train crisis managers and field crews. It allows the analysis and the evaluation of
complex incidents, their impact on the population and contingency scenarios (Boin and Schaap, 2010).
Projects such as KOKOS (KOKOS, 2015) invent methods, technical concepts and IT tools to involve the active
participation of the general public for self-help activities and communities integrated in processes of authorities.
To enable the interoperability between first responders and police authorities during crisis situations or disasters,
the project SECTOR (SECTOR, 2014) aims at establishing a secure European common information space
(CIS). The CIS should give continuous and shared access to all necessary data and information, besides the use
of collaboration process models to support coordination and cooperation between the organisations.
Another information management tool for the response phase is Disaster LAN
(https://www.buffalocomputergraphics.com/dlan/emergencymanager). This web-based task, mission and
resource management system for use in emergency operation provides tracking and reporting tools, emergency
communication tools and visual situational awareness tools (e.g. Monitor and respond to Twitter and social
media feeds, displays traffic cameras and video feeds, displays current weather, forecasts, warnings, and
animated radars, includes user-to-user and system-to-system messaging…) to guarantee efficient team work. It
755
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
is accessible for desktop and mobile devices. The tool seems to have a wide acceptance but appears to be only
US-based and therefore not useful for European crisis managers (Boin and Schaap, 2010).
Besides, the Foundation SAHANA (https://sahanafoundation.org/) provides free and open source disaster
management software. The offered information management solutions shall enable practitioners to better
prepare and respond to disasters. It is a web based collaboration tool that addresses the common coordination
problems during a disaster that help solve concrete problems (e.g. finding missing people, managing aid,
managing volunteers, tracking camps) and ensures efficiencies in disaster response coordination between several
participants (governments, aid organizations, the civil society (NGOs). The SAHANA Eden software offers a
flexible, modular platform rapidly deploying information management systems separated to five categories
(emergency, health, logistics, population and collaboration) with sub-categories for disaster management and
humanitarian use cases. Except that the system is web-based and vulnerable if connection is not available, the
system seems to be an appropriate tool for coordination during humanitarian aid operations and it addresses the
problems of a common coordination during disaster (Boin and Schaap, 2010).
EmerGeo Solutions (http://www.emergeo.com/) offers a worldwide emergency and crisis management software
and services that include two integrated products: Fusionpoint and EmerGeo Maping to use for planning,
training and response. It is a subsidiary of Sai Infosystems Ltd. and the product is already used in market.
Fusionpoint is a web-based software application that supports communication, collaboration and coordination
within and between emergency operation centers (EOCs) and the field. EmerGeo provides a Common
Operating Picture (mCOP™) technology, based on open standards, reliable world-class risk management
technology and professional services that are proven effective in helping government and industry to mitigate,
prepare for, respond to and recover from emergencies, natural disasters, acts of terrorism, and planned events
(Morrow ad Bachoura, 2013).
Swedish Decision Support System RIB is a system for prevention and emergency management. It includes the
computer system LUPP which is a tool for decision-making and follow-up of emergency response operations.
Since LUPP is a partial system within the support system, it implies the necessity for exchange of data between
different systems (Boin and Schaap, 2010).
Use of social media for crisis management: Ushahidi (https://www.ushahidi.com) and SensePlace2
(https://www.geovista.psu.edu/SensePlace2/): During the Haiti Earthquake (2010), the downloadable software
tool Ushahidi was first deployed. It enables people to submit eyewitness reports during a disaster that can then
be displayed onto a map in order to crowdsource data from people on the ground to aid relief efforts.
Responders can use the data to improve their situational awareness and make informed decisions. SensePlace 2
is another map-based web application that integrates multiple text sources (news, RSS, blog posts) that can then
be translated onto a map to allow emergency responders to easily filter through by place or time e.g. to analyse
changing issues and perspectives (Chan, 2013).
Despite the existence of the manifold tools, one major challenge regarding humanitarian supply chain
management systems is the fact that there is no tool being able to support all processes and tasks of logistics and
transport planning while fulfilling general requirements such as easy setup, good usability and low costs.
Aspects such as the analysis of the current logistics and traffic situation or an assessment of the infrastructure
are not considered by any of the projects (Detzer et al., 2016). Tools developed specifically for humanitarian
supply chains indeed mostly cover operational tasks leading to practitioners underestimating the tactical
dimension when making decisions. Consequently, the combined usage of several tools seems to be the best
option. But such a combination has to be well-considered as risks regarding data integration etc. are always
arising when using different systems (Blecken, 2010).
CONCENPTUALIZATION OF AN INTEGRETAD LOGISTICS AND TRANSPORT PLANNING APPROACH
Dealing with the consequences of a crisis requires various tasks and activities in a humanitarian supply chain.
Blecken (2010) has developed a reference task model to describe those and to point out how and where
information systems can support crisis manager professionals in a disaster situation. The reference task model
distinguishes between a strategic, tactical and operational planning horizon as well as functions assessment,
procurement, warehousing and transport. This braking down in distinct management levels and functional
dimensions results in twelve different areas each encompassing various tasks and in two additional activities -
reporting and operations support.
756
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
Figure 1. Reference task model framework (cf. Blecken, 2010)
Information systems can be used as a means of support for most of these tasks in several ways and this may
have a beneficial effect on enhancing humanitarian aid. In fact, the four components of agility –detection,
adaption, reactivity and effectiveness –define the way, how an information system can improve the execution of
the tasks described above (Bénaben, 2017).
xDetection: Regarding all functional dimensions, i.e. assessment, procurement, warehousing and
transport, monitoring capabilities of an information system can provide decision makers with relevant
data about the current situation, i.e. stock levels or condition of the road network. In this way, the task
to recognize and to keep the overview of deviations from expectations can be simplified.
xAdaption: Information systems can be applied to estimate the consequences of a certain strategy or
decision without actually implementing the strategy or decision, e.g. by using simulation tools. This
allows decision makers to select from a pool of suitable solutions and therefore enables to make
decisions on more than experience.
xReactivity: Information systems can accelerate the process of detecting deviations from the expected
situation and reacting accordingly by providing timely and valuable information and presenting this
information in a simple and comprehensible way.
xEffectiveness: Measuring the effectiveness of implemented strategies and decisions can be enhanced
by information systems that for example collect and evaluate data about all performed crisis activities.
Overall, an assessment of an operation’s success is easier when it is based on a comprehensive data
basis.
Available information system tools limit their scope in at least one dimension of the reference task model
framework (see Figure 1). While often several functionalities are covered, the planning horizon is typically only
considered on one level (namely strategic, tactical or operational). The majority of the provided tools
concentrate on the operational level and only a few tools address the strategic or tactical level. This leads to the
fact that decision makers in crisis situations usually underestimate tactical and strategic decision making and
focus more on operational actions. One disadvantage of such isolated tools is that they cannot benefit from
results or insights gained by other tools although decisions on different levels are dependent on each other. For
example, the set of possible solutions for an operational decision is restricted based on the decisions made on
strategic and tactical level. Such dependencies can be controlled, if information systems would be integrated on
different levels. Thus, each tool can benefit from the information and results of all embedded tools.
Especially for logistics planning and transport management, a merging of various tools would enable an
improved planning of resources’ prepositioning and operating of transport orders. Thereby, possible transport
routes as well as future traffic infrastructure conditions can be considered when deciding where to preposition
resources (e.g. food, tents, medicine…). Moreover, locations of available relief items and demands for aid
material can be considered into route planning. To unite tools of resource and route planning, it is essential to
exchange various relevant information that the other tool can benefit from. The logistics planning tool needs to
share information about warehouse locations, their inventory level of relief items, available vehicles and human
resources as well as location of emerging demand. If the transportation tool is capable of processing the received
information, it will be able to adapt suggested transport routes with regard to the given information in order to
respond to the depicted situation.
The transport management tool in turn needs to provide information about the road network’s condition,
757
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
expected travel times and most appropriate transport routes to the logistics planning tool. By combining such
information into e.g. simulation systems, the logistics planning tool becomes capable of suggesting better
decision alternatives.
Figure 2. Information exchange between logistics planning and transport management tools
In order to be able to foster the possible benefits, different conditions must be met. Moreover, a technical
integration of tools poses various risks, which need to be considered. For this, solutions to circumvent these
challenges have to be developed.
Technical risks and requirements
From a technical perspective, the aspect of data integrations needs to be treated carefully. Only when integrating
the different tools’ data cautiously these can be interpreted and applied correctly. To ensure coherent data
integration and valuable information exchange, consistent interfaces need to be programmed as well as suitable
data processing functionalities have to be provided. In general, good data quality is important to enhance user
acceptance.
Structural risks and requirements
Logistics planning and transport management tools operate on different planning horizon levels. Consequently,
the different management levels need to be respected to avoid an uncoordinated and insufficient interaction
between the tools. Moreover, risks of incorrectly shared information flows or overstimulation of information,
e.g. by processing irrelevant information, need to be compensated when integrating tools of different
management level. The overall aim is to consider the different levels and adapt the information flow to avoid
useless information.
Socio-economic risks and requirements
From a socio-economic perspective the user acceptance is a high risk. If the integrated tools are not accepted by
the user or do not address their actual needs, they will not achieve its objective and will not be capable of
offering any benefits. To prevent this risk, an easy and efficient handling of the integrated tools has to be
ensured.
Overall, a merging of a logistics planning and a transport management tool can be capable of closing the
identified gap in decision support regarding logistics and transport related issues. At the same time, it has to be
taken into account that such an integration comprises many risks and various requirements need to be
considered when implementing it. Overall, an integration of a logistics planning and a transportation tool can be
capable of closing the identified gap in decision support but such an integration offers many risks and various
requirements need to be considered when actually implementing it.
PROTOTYPICAL IMPLEMENTATION
Based on the requirements identified above we have designed an integrated solution which can be applied for
tactical as well as operational decision support within disaster relief operations covering logistics management
and transport planning tasks. Until now, the integration is limited to the four specific tools mentioned below, but
further integrations were considered during the design phase. Since the simulation-based logistics planning tool
758
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
supports various import and export functions using Open Database Connectivity (ODBC) other related
information systems (like warehouse management or enterprise resource planning systems) can be connected.
The integrated tool aims at supporting practitioners to have an overview on its structures, capabilities, capacities
and processes as well as to experiment with it during the preparedness and response phases. The integrated tool
can be applied by different stakeholders covering disaster relief organizations, crisis management authorities as
well as local and regional communities. However, the main target audience are relief organizations because the
tool allows an exact depiction of organization-specific setups (e.g. physical networks or definition of standard
operational procedures). The potential end users of the tool are the logistics and transportation responsibilities
from tactical and operational levels (like project programmer, heads of operations or warehouse managers). By
considering (global) assessment technologies, logistics planning methods as well transport management tools,
decision makers are supported in terms of data-based comparisons of alternative decisions, i.e. the integrated
tool is not aiming to tell practitioners what the optimal solution is, but it will demonstrate most probable
consequences of individual solutions made based on the experiences of the decision makers.
In the following paragraph, we first describe which technologies were integrated in the overall solution and
how. In the second part, we describe one particular experiment within the DRIVER (2014) project executed
with the Federal Agency for Technical Relief (THW). Within the experiment, we have modeled the THW
network and practices and simulated the relief operation based on a flooding in the city of Magdeburg in
Germany in the year 2013. Thus, the scope of the evaluation is restricted to a governmental crisis management
organization and the transferability to international relief organizations is limited. However, grounding the
simulation model with the reference task model developed with over 30 relief organizations promise to be
applicable to other organizations like international NGOs or UN organizations. Several of the involved tools
were already applied at other organizations like a Red Cross Society or an international operating NGO (see
e. g. Widera and Hellingrath 2011, 2016).
The integrated tool suite consists of four components: (1) KeepOperational, (2) U-Fly/3K, (3) ZKI-Tool and (4)
HumLog:
(1) KeepOperational is a web-portal developed by the German Aerospace Center (DLR). KeepOperational
visualizes the current traffic situation using different traffic sources. The traffic data can be used as
basis to simulate and predict traffic and for supporting the decision process for traffic management
actions (e.g. routing) in case of an incident. KeepOperational also involves SUMO –a microscopic and
open source road traffic simulation. It provides transport and traffic information for emergency services
(e.g. current traffic situation, routing advice, traffic prediction, scenario modelling, Presentation of the
timely reachability in dependency of the current traffic situation, display of aerial images).
(2) U-Fly/3K is a ground control station (GCS) for Remotely Piloted Aircraft (RPV). The capabilities
include mission planning and evaluation for single RPAS or swarm formations. U-Fly receives aerial
sensor data, processes and evaluates sensor data, and dynamically adapts RPAS missions to newly
received information. The 3K camera system is integrated into the RPV D-CODE and sends down
georeferenced images and derived image products to the GCS. Provision of airborne imagery data.
Information on damaged infrastructure can be extracted from the gathered data and traffic data can also
be extracted.
(3) The ZKI-Tool is provided by the Center for Satellite Based Crisis Information (ZKI) at the German
Remotes Sensing Data Center (DFD) at DLR. ZKI provides a 24/7 service for the rapid provision,
processing and analysis of satellite imagery during natural and environmental disasters, for
humanitarian relief activities and civil security issues worldwide. The tool provides flood impact
information, e.g. water masks, or information about affected infrastructures derived from aerial and
satellite imagery; it is used as an input for tools on traffic management and logistics support (e.g.
KeepOperational). Additionally, it provides 3-D visualization and GeoPDF for crisis management
organizations.
(4) HumLog itself is a tool suite containing three modules: (i) HumlogEM is a modelling tool able to
support various modelling languages (see e.g. Widera et al. 2013). It can be used for the application of
reference models as well as for model reporting and pattern search. (ii) HumLogSIM is an AnyLogic-
based simulation environment allowing a multi-method (discrete event, agent based, system dynamics)
simulation. The network structure and the logistics processes of relief chains are mapped, resulting in
organization-specific process models considering all relevant organization levels. The process models
will be then used to adjust the existing humanitarian logistics simulation environment in the simulation
model in AnyLogic. Appropriate parameter variations of the simulation model are then executed within
a simulation study which is then analyzed by utilizing the (iii) humanitarian logistics performance
measurement system HumLogBSC (see e.g. Widera et al. 2015, Widera and Hellingrath 2016).
The tool suite was applied in the following procedure: ZKI extracted crisis information from satellite and aerial
759
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
imagery (e.g. flood layers, flood impacts) and provided 2-D and interactive 3-D cartographic solutions (map
products and video animations) for the THW volunteers. By these means, the decision making processes were
supported by an increased situational awareness as well as a damage and needs assessment. Additionally, the
extracted crisis situation information was shared with KeepOperational. An airplane flew in advance of the
experiment over the affected area to collect airborne imagery. The imageries and extracted traffic and crisis
information were provided to KeepOperational. KeepOperational uses the provided information of U-Fly/3k and
ZKI-Products as well as traffic data information from other sources to recommend route options for emergency
vehicles by considering current traffic infrastructure, information of the traffic situation and infrastructure or a
traffic prediction & simulation. The gathered traffic and routing information was finally integrated into the
simulation environment HumLogSIM which evaluates different scenarios and network settings based on the
stored process models (e.g. material flow calculation, procurement analysis, scheduling, bottleneck analysis,
cascading effects). The architecture of the tool suite is depicted in Figure 3.
Scenario / Use Cases
Public Entities
(THW, MSB, DRK, …)
HumLog
KeepOperational
U-Fly/3k ZKI
Process chain and Tools
Evaluation
Crisis situation information (e.g. flooding map
Traffic and routing information
Traffic & crisis situation information
Modeling and Visualisation
Figure 3. Toolsuite Architecture
The data of the simulated scenario were based on the existing THW network (e.g. standard operational
procedures or location and vehicle information), recorded data (e.g. satellite imagery, aerial imagery) and event
logs (e.g. scheduling of volunteers, demand orders). After continuous rainfall over several days the major rivers
and its tributaries of Southern and Eastern Germany have reached their banks and became in danger of flooding
adjacent areas. The city expected the prospect of a major flooding of large parts of the city area and has started
emergency preparations for the event. The civil protection agency identified the endangered areas and affected
population as well as the critical infrastructure of the city. The experiment was executed as a purely table top
exercises, i.e. completely simulation-based.
We have recorded and analyzed quantitative and qualitative data. The generated results were presented to the
professional responders, who expressed high interest in the provided solutions and confirmed an added value to
their legacy systems and procedures. A detailed overview of the evaluation approach and the results can be
found in Detzer et al. (2016). In particular, the participants perceived the provided solutions as a suitable
solution for the actual transport and logistics tasks and processes. Besides, several improvements were identified
with regard to technical and functional aspects of the tool suite. It was stated by the practitioners that the
proposed solutions are beneficial for the THW volunteers regarding certain conditions:
operforming operation in unknown areas,
operforming tasks with considerably calculation effort,
operforming nationwide operations,
operforming complex tasks with many alternative decision choices.
Regarding the dimensions of agility, the evaluation showed that the following improvements could be
established by the tool suite:
xDetection: Especially the KeepOperational feature of displaying the current traffic situation has been
perceived as very good. In combination with HumLogSim’s route planning feature, it has been capable
of improving situation awareness (cf. Figure 4).
xAdaption: It was realized that the solutions provide a relevant data-driven basis for decision making,
because fast allocation and evaluation of alternatives was supported. Especially regarding
760
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
transportation tasks, the volunteers perceived this as useful. Nonetheless, it has been stated that the
usage of the provided solutions benefits at the level “management and communication”. During the
experiment it was noticed, that the benefits of the solutions below this level are limited for the
volunteers.
xReactivity: In order to understand the impact of the tool suite, a control group was carrying out the
same tasks in a traditional way. At the beginning of the experiment, it was noticed that the tool group
compared to the control group needed more time to complete the given tasks, especially in performing
easy and common tasks they are used to. The main reason for the time delay is that the volunteers
received only a very rough coaching on how to use the tool suite. The performance in terms of time
and accuracy of the results increased with an increase of the difficulty of the tasks.
xEffectiveness: The tool suite collects data that allow tracking of each transport and assessing
measurements such as speed, capacity utilization etc. These features have e.g. also been used to
evaluate the tool group’s performance during the experiment.
Overall, the results from the qualitative data approve the results from the quantitative performance analysis. The
professional responders declared a potential benefit for the tool suite and expressed their interest in applying it
in their daily work. The tool suite has been approved as suitable to be used in an operation by the volunteers (cf.
Figure 4).
However, several limitations were identified. The results’ regarding usability differs between the proposed
solutions. The operation of HumLog was not seen as easy and the use was assessed between good and partly.
The use of ZKI-Products and KeepOperational were evaluated between very good and good (cf. Figure 5). This
difference in the usability is justified by the fact, that, HumLog is not designed for field staff but for a yet not
existing THW system operator. The “tool” needs thus to be understood rather as a socio-technical change of the
current standard procedure. The suitability and relevance was confirmed for all components. Almost all
questionnaires stated that they could manage the tasks more easily, that they could finish the tasks faster due to a
better situation awareness. An interesting observation has been made regarding the (technical) connection of the
tool components. A deeper integration was expressed here. Another limitation was identified in an additional
mobile interface of all deployed components. Finally, the handling of the solutions and the data visualization
and accessibility should be further ensured improved.
Figure 4. Validation of Suitability
761
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
Figure 5. Validation of Usability
CONCLUSION
Logistics and transport planning can be described as a main driver for effective relief operations. In decision
makers’practices both tasks are often perceived as directly inter-related and its execution is based on intuition
and individual experiences of the involved staff. However, the available information systems address those
needs inadequately because they tend to focus only on particular tasks for a particular sequence. Besides, those
tools mostly use optimization approaches and seem to relieve final decisions of the practitioners. These
circumstances lead to a poor acceptance of promising innovations and improvements of those truly challenging
tasks in relief operations. We proposed and evaluated an integrated approach of logistics and transport planning
which connects planning horizons and the inter-related tasks. The tabletop exercise revealed a relatively high
acceptance by the practitioners and it has shown a direct impact on the logistics performance. Additionally, a
couple of limitations were identified, especially in terms of embedding new practices into standard operational
procedures, the usability of several tool suite components as well as their integration. We identified specific
adjustments of the developed tool suite with regard to the given feedback. The next steps must consider further
iterations of development and testing. For the purpose of testing it will become necessary to deeper investigate
appropriate application scenarios as well as to analyze and evaluate new operational practices in close loops
with the practitioner organizations. These lessons learned are based on the confirmed assumption that only a
socio-technical approach promises to unfold the available benefits to improve crises management practices.
ACKNOWLEDGMENTS
The research leading to these results has received funding from the European Union Seventh Framework
Programme (FP7/2007- 2013) under grant agreement n° 607798. Moreover, the experiment presented in this
paper would not have been possible without active support of the THW employees and volunteers. Finally, we
thank the DRIVER “Logistics Management and Transportation” experiment team that has been working
together for months in order to prepare, conduct and finally analyze the experiment. (*)
REFERENCES
Altay, N. and Green, W. (2006) OR/MS research in disaster operations management, European Journal of
Operational Research, 175, 1, 475-493.
Ben Othmann, S., Zoghlami, N., Hammadi, S. and Zgaya, H. (2014) Adaptive Collaborative Agent-based
System for Crisis Management, International Joint Conferences on Web Intelligence (WI) and Intelligent
Agent Technologies (IAT), Warsaw, Poland.
Ben Othmann, S., Zgaya, H., Dotoli, M. and Hammadi, S. (2017) An agent-based Decision Support System for
resources’ scheduling in Emergency Supply Chains, Control Engineering Practice, 59, 27-43.
Benaben, F., Vernadat, F. B. (2017) Information System agility to support collaboratie organisations, Enterprise
Information Systems, 11, 4, 470-473.
Bevere, L., Seiler, T., Zimmerli, P. and Feyen, H. (2013) Natural catastrophes and man-made disasters in 2012.
A year of extreme weather events in the US. Swiss Re (sigma, 2/2013).
Blecken, A. (2010) Humanitarian Logistics: Modelling Supply Chain Processes of Humanitarian Organisations,
762
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
Haupt Verlag AG.
Boin, A. and Schaap, M. (2010) State of the art Crisis Management Tools, INDIGO Crisis Management
Solutions. http://indigo.diginext.fr/EN/Documents/I4.4.4%20INDIGO-
StateOfTheArt%20CM%20tools%20V1.1.pdf
Chan, J. C. (2013) The role of social media in crisis preparedness, response and recovery.
http://www.oecd.org/governance/risk/The%20role%20of%20Social%20media%20in%20crisis%20prepared
ness,%20response%20and%20recovery.pdf
Charles, A., Lauras, M., van Wassenhove, L. N. and Dupont, L. (2016) Designing an efficient humanitarian
supply network, Journal of Operations Management, 47-48, 58-70.
CRISMA (2015) Project CRISMA: “Modelling crisis management for improved action and preparedness”
(project period 2012 –2015). Available online: http://www.crismaproject.eu/
D’Uffizi, A., Simonetti, M., Stecca, G. and Confessore, G. (2015) A Simulation Study of Logistics for Disaster
Relief Operations, Procedia CIRP, 33, 157-162.
Degener, P., Gösling, H. and Geldermann, J. (2013) Decision support for the location planning in disaster areas
using multi-criteria methods. Proceedings of the 10th International Conference on Information Systems for
Crisis Response and Management (ISCRAM).
Detzer, S., Gurczik, G., Widera, A. and Nitschke, A. (2016) Assessment of Logistics and Traffic Management
Tool Suites for Crisis Management, European Transport Conference.
DRIVER (2014) FP7 EU Project “DRiving InnoVation in crisis management for European Resilience” (project
period 2014 –2018). Available online: http://www.driver-project.eu.
ELITE (2014) Project ELITE: “Elicit to Learn crucial post-crisis lessons” (project period: 2013 – 2014).
Available online: http://www.elite-eu.org/index.html
Fallucchi, F., Tarquini, M. and de Luca, E. W. (2016) Knowledge Management for the Support of Logistics
during Humanitarian Assistance and Disaster Relief, In: Díaz, P., Bellamine, N., Saoud, B., Dugdale, J. and
Hanachi, C., Information Systems for Crisis Response and Management in Mediterranean Countries, 265.
Fikar, C., Gronalt, M. and Hirsch, P. (2016a) A decision support system for coordinated disaster relief
distribution, Expert Systems with Applications, 57, 104-116.
Fikar, C., Hirsch, P., Posset, M. and Gronalt, M. (2016b) Impact of transalpine rail network disruptions. A study
of the Brenner Pass, Journal of Transport Geography, 54, 122-131.
FORTRESS (2014) Project FORTRESS: “Foresight Tools for Responding to cascading effects in a crisis”
(project period: 2014 –2017). Available online: http://fortress-project.eu/
Garrido, R. A., Lamas, P. and Francisco, J. (2015) A stochastic programming approach for floods emergency
logistics, Transportation Research Part E: Logistics and Transportation Review, 75, 18-31.
Irohara, T., Kuo, Y.-H., Leung, J, M. Y. (2013) From Preparedness to Recovery: A Tri-Level Programming
Model for Disaster Relief Planning. In: Hutchison, D., Kanade, T., Kittler, J., Kleinberg, J. M., Mattern, F.,
Mitchell, J. C. et al. (eds.), Computational Logistics, 8197, Springer Berlin Heidelberg (Lecture Notes in
Computer Science), 213–228.
Kelle, P., Schneider, H., Yi, H. (2014) Decision alternatives between expected cost minimization and worst case
scenario in emergency supply –Second revision, International Journal of Production Economics, 157, 250–
260.
KOKOS (2015) Project KOKOS: „Unterstützung der Kooperation mit freiwilligen Helfern in komplexen
Schadenslagen (KOKOS)“ (project period: 2015 – 2018). Available online: http://kokos.wineme.fb5.uni-
siegen.de/
Kuo, Y.-H., Leung, J. M. Y., Meng, H. M. and Tsoi, K. K. F. (2015) A Real-Time Decision Support Tool for
Disaster Response: A Mathematical Programming Approach, International Congress on Big Data, New
York City, USA.
Locra, A., Celik, M., Özlen, E. and Keskinocak, P. (2016) An Optimization-based Decision-Support Tool for
Post-Disaster Debris Operations, Production and Operations Management.
Morrow, M. and Bachoura, M. (2013) Crisis Management & Situational Awareness System. An integrated
approach. http://www.emergeo.com/wp-content/uploads/2013/03/Fusionpoint-Case-Study.pdf.
Ransikarbum, K. and Mason, S. J. (2016) Goal programming-based post-disaster decision making for integrated
relief distribution and early-stage network restoration, International Journal of Production Economics 182,
324–341.
763
Widera et al.
Integrated Logistics and Transport Planning in
Disaster Relief Operations
WiPe Paper –Logistics and Supply-Chain
Proceedings of the 14th ISCRAM Conference –Albi, France, May 2017
Tina Comes, Frédérick Bénaben, Chihab Hanachi, Matthieu Lauras, Aurélie Montarnal, eds.
SECTOR (2014) Project SECTOR: “Secure European Common Information Space for the Interoperability of
First Responders and Police Authorities” (project start: 2014). Available online: http://www.fp7-sector.eu/.
Semini, M. and Fauske, H. (2006) Applications of discrete-event simulation to support manufacturing logistics
decision-making. A surey., Proceedings of the 2006 Winter Simulation Conference
Terzi, S. and Cavalieri, S. (2004) Simulation in the supply chain context. A survey, Computers in Industry, 54,
3-16.
Widera, A. and Hellingrath, B. (2011) Improving Humanitarian Logistics —Towards a Tool-based Process
Modeling Approach. In Proceedings of the LM 2011, Bamberg, 273–295.
Widera, A. and Hellingrath, B. (2016) Making Performance Measurement Work in Humanitarian Logistics. The
Case of an IT-supported Balanced Scorecard. In Haavisto, I., Kovacs, G., & Spens, K. M. (Eds.), Supply
Chain Management for Humanitarians. Tools for Practice (1st ed., pp. 339–352). Kogan Page.
Widera, A., Hellingrath, B., & Bubbich, C. (2015). Humanitarian Logistics Dashboards —A Design-related
Requirements Analysis. In Proceedings of the IEEE Global Humanitarian Technology Conference, Seattle,
USA, 92–99.
Widera, A., Dietrich, H.-A., Hellingrath, B., & Becker, J. (2013). Understanding Humanitarian Supply Chains
—Developing an Integrated Process Analysis Toolkit. In Proceedings of the International Conference on
Information Systems for Crisis Response and Management (ISCRAM) 2013, Baden-Baden, Germany, 210–
219.
764