Community Research and Development Information Service - CORDIS

FP7

CRISMA Report Summary

Project reference: 284552
Funded under: FP7-SECURITY

Final Report Summary - CRISMA (Modelling crisis management for improved action and preparedness)

Executive Summary:
The CRISMA (Modelling crisis management for improved action and preparedness) project was an integration research project funded by European Union 7th framework programme. This three and a half year project started March 1, 2012 and had a total budget of 14,4 million Euro of which the funding from the European Commission was 10,1 million Euro. The overall aim of the project was to provide a simulation-based decision support system for modelling crisis management in both natural and man-made crisis, specially those complex ones with low probability and high impact.

Instead of developing individual crisis management solutions from scratch, a generic software framework – CRISMA Framework – has been developed in CRISMA. The CRISMA Framework is composed of a range of different software components, models and supporting tools that can be combined in various ways to form different simulation applications for different kinds of crisis domains.

With the CRISMA Framework, crisis managers and other decision-makers are offered possibilities to combine models, data and expertise originating in many different sources for creating a wider perception of crisis scenarios and better awareness of alternative preparedness, response and mitigation actions. Moreover, CRISMA scenario comparison and visualisation tools can be utilised to improve cooperation between many organisations as well as communication with other stakeholders and the public.

The feasibility of the Framework, the developed software components, and the underlying decision-support concepts have been tested and validated in five pilot sites, representing a rich set of illustrative cases to demonstrate the benefits of CRISMA for different end-users in various crisis situations.

Project Context and Objectives:
In recent years disasters have significantly affected populations and their living environments globally. Decision makers at different levels appeared to be unprepared for responding adequately to new magnitudes of crisis management demands.

Natural and human-made disasters are not events, but processes, in which previous historical responses to events contribute heavily to the degree of preparedness and the extent and nature of impacts and the recovery from disasters offers important opportunities to address underlying causes, problems and institutional incapacities. The extent of the impact of a disaster is closely related to the capacity of institutions and the public to learn and adjust from previous experiences, and on the pre-disaster performance and capacity of infrastructure – transport systems, electricity and water supply. The more that policy makers and citizens understand the crisis evolvement and likely impacts, the more they will prepare to increasing risks. It is apparent that policy level decision making plays a large role in the degree of preparedness, and that the complexity of the situation and impacts call for multi-sectoral coordination between authorities and other stakeholders.

Crisis management is a key sector under the overall security business activities, and the market is growing along emerging threats and expansive consequences of hazards. Hardly any mission could succeed without private companies and contractors.

Services delivered by private companies include medical care, logistics, transportation and information systems and infrastructure (Kupi 2010). Wide spectrum of existing national information systems and command and control tools require high flexibility and open integration interfaces. As the threats are changing, decision support systems should include a provision not only for existing but also future decision-making situations. It also important to look the whole continuum and identify flexible business models, solutions and technologies for both the crisis management operations, preparedness planning, as well as for the rebuilding process. In the area of crisis management private companies cooperate with local authorities, governments, international organisations and civil-society actors. Public-private partnerships therefore play an important role in the crisis management business and also in building up preparedness. Information technology based solutions, which deliver shared situational awareness and open interfaces for integration, and thus support timely decision making and the use of GUI information, are relevant though the whole crisis cycle from prevention and preparedness beyond the acute rescue actions. While the CRISMA project focused on planning for preparedness and response, the results benefit other steps of the crisis cycle too.

CRISMA Integrated Project aimed to support both public and private crisis manager and decision-makers in short and long term planning as well as training and reviewing of crisis preparedness and response organisations, infrastructure and personnel. The main objective was to provide a modular and user-friendly, simulation-based system for decision support in crisis management during the real-life activities in planning, training and operations. The focus was on large-scale crisis scenarios with immediate and extended human, societal, structural and economic, often irreversible, consequences and impacts. Typically, these crisis scenarios cannot be managed alone with regular emergency and first responder resources, but require multi-organisational and multi-national cooperation including humanitarian aid.

The CRISMA consortium planned to develop an integrated planning and decision support tool set that facilitates simulation and modelling of realistic crisis scenarios with possible cascading and multi-risk effects, potential preparation and response actions, and the impact of crisis depending on both the external factors driving the crisis development and the various actions of the crisis management team. The CRISMA tool set enables decision makers and crisis managers to: (1) model possible multi-sectoral crisis scenarios and assess the consequences of an incident, (2) simulate possible impacts resulting from alternative actions, (3) support strategic decisions on capabilities, related investments, reserves, inventories (4) optimise the deployment of resources dedicated to crisis response in-line with the evolvement of a crisis, and (5) to improve action plans for preparedness and response phases of the crisis management.

Instead of developing individual crisis management solutions from scratch, a generic software framework – CRISMA Framework – has been developed in CRISMA. The CRISMA Framework is composed of a range of different software components, models and supporting tools that can be combined in various ways to form different simulation applications for different kinds of crisis domains.

With the CRISMA Framework, crisis managers and other decision-makers are offered possibilities to combine models, data and expertise originating in many different sources for creating a wider perception of crisis scenarios and better awareness of alternative preparedness, response and mitigation actions. Moreover, CRISMA scenario comparison and visualisation tools can be utilised to improve cooperation between many organisations as well as communication with other stakeholders and the public.

The feasibility of the Framework, the developed software components, and the underlying decision-support concepts have been tested and validated in five different simulation scenarios of crisis management. The corresponding simulation scenarios have been implemented in the respective CRISMA reference applications developed for the following domains:
• Nordic winter storm domain;
• Coastal submersion domain;
• Accidental pollution domain;
• Earthquake and forest fire domains;
• Resource management training and resource planning domains.

As a result of the testing and validation in five different domains, we can conclude that different crisis management applications built around the CRISMA Framework can support numerous types of decisions. In general, decision-support functionalities of the CRISMA Framework enable the end users to analyse and compare different simulation scenarios in order to identify sound and efficient mitigation strategies, which can either be included in preparedness plans or used to train decision-makers and other stakeholders. In particular, we established that the five reference applications that were used for experimentation and testing in CRISMA help decision-makers in the following ways:
• by allowing experimentation with different crisis management strategies and decisions;
• by supporting decision-makers in long-term planning;
• by enabling simulation of natural disasters in a multi-risk framework including cascading effects;
• by facilitating resource management and planning.

The generic CRISMA Framework is based on the methodology for simulation-based decision support worked out in the project. This decision-support methodology with the related software emphasise the fact that criteria and ranking functions represent opinions rather than facts. Both the criteria and ranking functions are highly situation-dependent and different stakeholders are likely to disagree on definitions and relative importance of different criteria. The end users are therefore encouraged to define several sets of criteria and ranking functions and compare the outcomes of applying them in different alternative simulation scenarios.

Project Results:
1 PROJECT STRUCTURE

The overall strategy of the work plan of CRISMA was to integrate different crisis management tools and incident propagation models and substantiate them with real and reference scenarios. On this fundament, simulation and modelling facilities were aimed to allow describing situations better and simulating alternatives, which help in getting a better understanding of hazards and measures to be taken.

The structure of the CRISMA project consisted of seven sub-projects:
• SP1 Cross-Sub-Project Coordination and Validation
• SP2 Scenarios, Requirements and Criteria for Crisis Management Modelling
• SP3 Integrated Crisis Modelling System
• SP4 Models for Multi-Sectoral Consequences
• SP5 Experimentation and Testing
• SP6 Dissemination and Exploitation
• SP7 Project Management

In CRISMA work-plan, high scientific and technical outcomes have been ensured through close interdependencies between SPs 2-5, and through constant dialogue between technical developers and end-users. CRISMA has been carried out in close cooperation with the end-user partners who have wide experience on crisis management in complex situations including national disasters and global response activities. The work has been also guided and supported by the End-User Advisory Board as well as other external end-users throughout the project lifetime. The outcome of CRISMA has gone through iterative end-user testing and final demonstrations in five different pilot configurations. The aim has been to build up acceptance in the wide end-user community in order to support the exploitation of project outcome and reduce the time-to-market after the project.



Figure 1. CRISMA project structure and strategy

2 SCENARIOS, REQUIREMENTS AND CRITERIA FOR CRISIS MANAGEMENT MODELLING (SP2)

SP2 was aimed to structure and pre-process existing knowledge on crisis management for its modelling, and to clarify and specify user requirements and expectations towards CRISMA. For this purpose, the following aspects were specifically covered:
• Review of available knowledge and capability bricks from partners, EU projects and literature
• Specifying sample and reference crisis scenarios in a common methodology (especially to SP3 and SP5)
• Define user requirements and use cases for the crisis management modelling system (especially to SP1, SP3 and SP5 as well as SP6)
• Elaborate general quantified criteria to be used in the crisis management models and tools (especially to SP3 and SP4)

Main focus was at understanding the needs of the CRISMA end-users and transferring the needs of the CRISMA end-users to the structure developed for the CRISMA solutions. The targeted end-users were those crisis management practitioners in charge of training and planning activities related to medium and large scale crisis scenarios.

The key outputs from SP2 included:
• Architectural design decisions that provided the decision rationale for the selected CRISMA architecture and explained technological decisions for the CRISMA Framework and applications.
• Detailed reference scenarios in a format that supports the work of SP3 and SP5 teams by the presentation of WSs, transition points and used models for reference scenarios.
• Use cases and interaction concepts that defined storyboards for the interaction with the CRISMA system in relevant use cases and for the interaction with new CRISMA models. Additionally, a general user interaction concept for the planning use cases was elaborated.
• KPI elaboration in a manner that depends on defined indicators that can be used in decision support to reduce the complexity of crisis management scenarios to consistent and comparable key aspects. Indicators aim to describe and quantify several aspect of e.g. hazards, vulnerabilities and impacts, capacities and resources, and economic impacts.

For the basis of the model and framework development, existing legacy models were mapped and analysed for different hazards, vulnerabilities, losses, impacts, and planning of capacity, capability and resources for rescue and mitigation actions (Cabal (Ed.) 2012). This extensive analysis supported the more detailed focusing of the work into those features that the end-user community were missing on the field. The inventory of desired data structures as well as interface specifications further guided the technical development in SP4 but also in SP3.


3 INTEGRATED CRISIS MODELLING SYSTEM AND MODELS FOR MULTI-SECTORAL CONSEQUENCES

The overall aim was to develop an easy-to-use planning, prediction, decision support and crisis simulation framework, for the use in multi sectoral crisis management applications based on a what-if scenario execution and management environment, which shall help to optimise the use of various resources to mitigate crisis events. The scope was also to provide the necessary tools in support of the crisis management system implemented in the CRISMA project. The scope covered the aim to simulate consequences in multiple sectors and systems (e.g. built environment, roads and transport systems, lifelines, society, business and production systems), by considering either single triggering events or cascading effects, time dependant vulnerability of the elements at risk, or systemic vulnerability.

These objectives have been reached through:
• CRISMA Decision support concept
• CRISMA Framework
• CRISMA crisis management models
• Building Blocks of the CRISMA Framework

Instead of developing individual crisis management solutions from scratch, a generic software framework – CRISMA Framework – has been developed in CRISMA. The CRISMA Framework is composed of a range of different software components, models and supporting tools that can be connected together to form different simulation applications for all kinds of crisis domains.

With the CRISMA software Framework, crisis managers and other decision-makers are offered possibilities to combine models, data and expertise originating in many different sources, for creating a wider perception of crisis scenarios and alternative preparedness, response and mitigation actions. Moreover, CRISMA scenario comparison and visualisation tools can be utilised to improve multi-organisational cooperation and communication with other stakeholders and the public.

The feasibility of the CRISMA Framework, the developed software components and the underlying decision-support concept have been tested and validated in five real-life cases representing different crisis management contexts. Different crisis management applications built around the CRISMA Framework can support numerous different types of decisions. In general, it enables the users to analyse and compare different scenarios in order to identify sound and efficient mitigation strategies, all of which can either be included in preparedness plans or used to train decision-makers and other stakeholders.

The generic CRISMA software Framework is based on a simulation and decision support concept. This decision-support methodology with the related software emphasise the fact that criteria and ranking functions represent opinions rather than facts. They are highly situation dependent and different stakeholders are likely to disagree on definitions and relative importance of different criteria. Users are therefore encouraged to define several sets of criteria and ranking functions and compare the outcomes.

3.1 CRISMA Decision support concept

CRISMA Decision support concept consist of elements like world states, indicators and decision support principles that translate real-life crisis challenges into the CRISMA Framework and its simulation and modelling environment. These elements and their role in CRISMA decision support concept is presented below.

World states (WS)
The situation of the world at a given time during a crisis management scenario is represented in the CRISMA system as a world state (WS), which contains a structured collection of data records – Objects of Interest (OOI). The scenario evolves in terms of the CRISMA world by transitions from one WS to the next WS, while the content of OOIs is updated by simulations and services applied at the transition. (Engelbach, W. (Ed.), 2014)

The initial WS (Figure 2) is specified by an initial collection of OOIs representing the inventory data, meteorological data etc. provided by the CRISMA user and/or retrieved through dedicated applications and web services that are interfaced with the CRISMA system.

Starting from the WS of interest the CRISMA user can simulate the evolvement of generic crisis phases (e.g. preparedness, response) either by conducting predefined simulations from the current WS or through direct manipulation of the current WS data. The WS transition is conducted by one or several simulation models attached to the transition that take selected WS data as input values and produce new WS data as output values. Multiple evolvement traces are acceptable as the WS can also be updated and changed by a CRISMA user to define alternative scenarios. This is represented in Figure 2 by the manipulation feature. The knob drawn within the “Manipulation” tag represents the user’s interaction with the system either by directly manipulating the WS data, or by tuning the control parameters of a simulation model. Moreover, in a WS, indicators, criteria and costs can be calculated to give representative and quantitative information for the analysis and assessment of the current WS. This analysis supports decisions related to the selection of subsequent actions including the generation of alternative scenarios.



Figure 2. World state (WS) evolution in CRISMA

Time is one of the key WS parameters, which may, but does not always have to be simulated in WS transitions. In the simplest case the time can be introduced by a partial order of WSs (in a given timescale), where the next state is considered to appear later in time and simulated durations of WS transitions can vary. Evolvement of real time can be captured by time stamps attached to WSs. Let us mention that this alternative depends on the timescale of simulations applied to transitions because there are (intentionally) no means for synchronization of simulation activities except through manipulating the WSs themselves. It is also possible to record time series for the transition period between two WSs. This can help to keep the number of WSs small, while still retaining the possibility to analyse the evolvement of time on a higher level of granularity.

Indicators
Indicators form the basis of decision support (DS) provided by CRISMA (Engelbach, et al, 2014; Engelbach (Ed.) 2014). Indicators in the CRISMA Framework characterise aspects of interest of the quantitative statuses of WSs within scenarios. An indicator is represented by a single scalar value attached to a CRISMA WS and can be, for example, aggregate values from the current WS by a function, e.g. average, minimum or maximum values in a given region for a given time. An indicator may be represented numerically or graphically as a spatial indicator, such as a color-coded map indicating, for example, the number of casualties in different areas.

Indicators may refer to hazard, vulnerability, impact or response aspects of crisis management. They can also describe the relationships such as the number of first responders with regard to the number of affected people, or the percentage of injured people. Even though indicators are neutral and purely descriptive, the selection of indicators is nevertheless a statement referring to which aspects of crisis management are relevant to consider. For the CRISMA Framework, it is recommended to select indicators for some general principles and specify related indicator functions. The main principle for selecting indicators for the CRISMA Framework is that any indicator should be expressed in measurable, replicable and reasonably easy to interpret units in the crisis management context.

A criterion is the qualitative counterpart of a quantitative indicator from a decision-making perspective. That criterion is constructed by a criterion function that defines the value-intervals for an indicator and assigns the “scale of satisfaction” to these intervals.

Decision-support principles
The overall idea of the CRISMA Decision-support concept and the CRISMA Framework is to (Taveter, K. (Ed.), 2015a):
(a) Let the decision-maker produce and use scenarios in support of his/her decisions
(b) Provide aggregated but representative information about the scenarios in the form of indicators at WSs
(c) Support the decision-maker in defining an explicit decision-making strategy; and
(d) Assist the decision-maker in comparing and ranking scenarios according to the decision-making strategy.

As described in Figure 3, the overall decision-support concept consists of: (1) an impact, mitigation or mission scenario represented by a set of consecutive WSs – a pre-specified record of information for decision-support; (2) indicator functions applicable to WSs; (3) a set of representative indicator values of the scenario (resulting from applying the indicator functions to the WSs of the scenario); (4) criteria functions that map indicator sets to (5) the level of satisfaction (decision criteria) on a normalised scale (0-1 or 0%-100%); and (6) the scoring function mapping a vector of criteria to a single scalar “score” value that can be used to sort scenarios and determine their rank among the elements of the scenario selection.



Figure 3. Overview of the decision-support concept of CRISMA

The decision-maker can use four types of information entities – impact mitigation or mission scenarios, scenario indicators, decision criteria, scores and scenario ranks – as a basis for his/her decisions and specify his/her individual decision-making strategy by the individual definition of criteria functions mapping indicators to the selected criteria. The decision-maker can tune decision-making also by assigning priorities to indicators as well as by applying their compositions.

In summary, the decision-maker can:
• Use indicators derived from scenario data (usually aggregated) to quickly assess and compare scenarios;
• Define a decision-making strategy by:
• Mapping performance indicators to decision criteria by defining for each indicator a function that maps indicator values to the corresponding level of satisfaction;
• Defining priorities (levels of importance) by assigning weights to indicators;
• Defining the level of Andness and Orness (Dujmović, 2007) to be considered when computing the rank of a scenario;
• Deal with a multi-criteria decision problem by obtaining a ranking of scenarios with respect to the decision-making strategy defined.

Ordered weighted averages as a mean for decision-support
Decision-making problems considering more than one criterion on the basis of, for example, impact scenarios require appropriate methods to assess the performance of specific scenarios. For the CRISMA decision-support concept the Ordered weighted averages method (OWA) (Yager, 1988; Yager, 1996; Zuccaro & Filomena, 1988) has been selected. OWA allows one to specify a particular multi-criteria decision-making strategy by defining the following properties of a good solution:
• Implement several decision-makers’ perspectives (multiple points of view);
• Make the decision-making strategy more explicit;
• Obtain a score/rank for each scenario;
• Let the decision-maker choose between different decision-making strategies (e.g., optimistic, neutral, pessimistic);
• Compare the results obtained under different strategies.

Using OWA, normalised indicator values (criteria) are multiplied with the corresponding levels of importance. The vector of weighted levels of satisfaction for all indicators is ordered according to their absolute values and weighted according to their position in the vector. In CRISMA these weights are defined manually by the decision-maker (Taveter, K. (Ed.) 2015a).

3.2 Decision-support functionalities

The CRISMA decision-support methodology and software emphasise the fact that criteria and ranking functions represent opinions rather than facts. They are highly situation dependent and different stakeholders are likely to disagree on the definitions and relative importance of different criteria. Users are therefore encouraged to define several sets of criteria and ranking functions and compare the outcomes.

CRISMA decision-support principles are depicted in Figure 4. From left to right, the amount of information is reduced from “complete world data” to a single number.



Figure 4. Summary of the CRISMA decision-support methodology (Taveter (Ed.) 2015b; Heikkilä et al (Eds) 2015)

By CRISMA, the following decision support functionalities are achieved:
• Viewing WSs enables to browse and view different scenarios that have been already created as well as browse the information recorded in any of the available WSs of these scenarios
• Executing simulations in order to discover alternative scenarios. Virtually every WS within any alternative scenario can be picked as the starting point of a simulation
• Applying mitigation options in the simulations in order to produce alternative “sub-plans”
• Investigating cascade effects to get insight into possible hazards that are initiated by the impacts of primary hazards as triggering events. It is implied that there is already a (simulated) primary hazard with a certain intensity available
• Analysing and comparing scenarios that allows to consider multiple scenarios at different abstraction levels, in order to find the ”best” solution under the current decision objectives. The abstraction levels of decision-support in analysing and comparing scenarios include:
o WS (provides insight of the information captured by a WS)
o Indicators of a WS
o Criteria (qualifies indicators by means of user-defined criteria functions)
o Multi-criteria analysis (for specification of user-defined decision strategies).

CRISMA enables building up various applications by combining available Building Blocks according to any set of required functionalities.

3.3 CRISMA Framework

The CRISMA Framework (Dihè (Ed.) 2015; Heikkilä et al (Eds) 2015) provides the software and tools that are necessary to develop concrete CRISMA applications. The CRISMA Framework consists of generic, composable and reusable Building Blocks (BB) that can be assembled into concrete applications for simulating crises and crisis management.

The CRISMA Framework provides infrastructure BBs, integration BBs, user interaction BBs, and crisis management models:
• Infrastructure BBs work in the background and constitute the backbone of a CRISMA application.
• Integration BBs support the integration of data and simulation models into a CRISMA application.
• User interaction BBs are graphical user-interface elements of a CRISMA application.
• Crisis Management Models are simulation models that cover different aspects of crisis management.

Figure 5 gives and overview on the Infrastructure Building Blocks of the CRISMA Framework, as identified, described and implemented in CRISMA.



Figure 5. Infrastructure Building Blocks (Taveter (Ed.) 2015b)

Figure 6 gives and overview on the Integration Building Blocks of the CRISMA Framework, as identified, described and implemented in CRISMA.



Figure 6. Integration Building Blocks (Taveter (Ed.) 2015b)

Figure 7 gives and overview on the User Interaction Building Blocks of the CRISMA Framework, as identified, described and implemented in CRISMA.



Figure 7. User Interaction Building Blocks (Taveter (Ed.) 2015b)


The complete list of all the software developed in CRISMA is presented in the CRISMA Catalogue (Havlik et al (Eds.) 2015) where all the Building Blocks are described in more detail.

3.3 CRISMA crisis management models

In the following sub-sections, the Crisis Management Models of the CRISMA Framework are introduced. Figure 8 gives an overview on them, as identified, described and implemented in CRISMA.



Figure 8. Crisis Management Models


Time Dependent Vulnerability Model
The Time Dependent Vulnerability (TDV) Model is a model for the assessment of time-dependent damage on elements at risk. The TDV Model allows performing consistent computation of time dependent damage by the use of suitably updated vulnerability functions. The TDV Model is “domain-independent” in the sense that the logic scheme is the same for different hazard domains (e.g. earthquake, flood, extreme weather etc.), but for using the model in order to compute time dependent losses (in terms of damages in the established damage scale) there is the need to suitably feed the model for each hazard-domain.

The logic for including TDV is shown in Figure 9.



Figure 9. The possible approaches for including time-dependent vulnerability in the flow of analysis (Polese & Zuccaro (Eds.) 2013)

For detailed description and application examples, please see the public deliverable D43.2 (Polese & Zuccaro (Eds.) 2014) and
https://crisma-cat.ait.ac.at/model/
Time%20Dependent%20Vulnerability%20model%20%28TDV%29


Cascading Effects Model
The Cascading Effect Model for dynamic scenario assessment calculates the probability of attainment of cascading events scenarios, given an initial triggering event, or estimates consequence paths given the occurrence of selected scenarios and considering alternative mitigation actions. Within the model, a set of scenarios and a Transition matrix are defined as the two fundamental pieces of information required to assess the effects of possible Cascading Effects.

Figure 10 illustrates the usage of the Cascading Effects Model in a CRISMA reference application.



Figure 10. Investigating cascading effects by means of the Cascading Effects Model

For a detailed description and application examples, please see the public deliverable D42.3 ( Almeida & Viegas (Eds.) 2014) and
https://crisma-cat.ait.ac.at/model/Cascading%20Effects%20Model

Economic Impacts Model
The Economic Impacts Model is a model for presenting economic impacts arising from crises (ex post performance) and assessing different mitigation proposals and their costs/benefits (ex ante planning). It is intended to be used in the preparedness phase of crisis management to support long-term strategic decision-making. The model uses data on alternative scenarios (e.g. base line scenario and crisis scenario after implementing a mitigation measure) in order to make the economic assessment. The assessment is done by determining economic losses of a crisis and costs and benefits linked to different mitigation investments.

The conceptual approach behind the economic impacts model in CRISMA is visualised in Figure 11.



Figure 11. Conceptual model for evaluation of economic impacts

For a detailed description and application examples, please see the public deliverable D44.2 (Engelbach (Ed.) 2014) and
https://crisma-cat.ait.ac.at/model/Economic%20impacts%20model

Population Exposure Model
The Population Exposure Model is a model for distributing population in spatial and temporal dimensions. It uses temporal and spatial proxies in order to disaggregate the population from administrative units to spatio-temporal grids. The outcome is used in CRISMA as basis for time-dependent exposure assessment and in further steps for evacuation and casualty modelling. Population data is usually available from census as totals per inhomogeneous spatial reference unit. For modelling population exposure, data is required that is independent from enumeration and administrative areas.

For detailed description and application examples, please see the public deliverable D43.2 (Polese & Zuccaro (Eds.) 2014) and
https://crisma-cat.ait.ac.at/model/Population%20exposure%20model

Resource Management Model
The Resource Management Models developed in the CRISMA project are build upon the OOI concept with different context dependent behavioural patterns for different crisis domains. Thus, there is no overall generic and all purpose Resource (OOI) Management Model, but a set of distinct models for different types of resources (e.g. ambulances, patients) and different situations. However, such domain and crisis specific Resource Management Models can be implemented on basis of the general Agent-Oriented Simulation Models Building Block.

For detailed description and application examples, please see
https://crisma-cat.ait.ac.at/model/Resource%20Management%20Model




3.4 CRISMA Building Blocks related to decision-support

In CRISMA there were a number of BBs developed in order to offer functionalities that are categorised as decision-support and form the CRISMA decision-support tool together with the main components overviewed above. In Figures 5, 6, 7 and 8 above, a full overview of the CRISMA Building Blocks serving decision-support is provided. More details on the specific Building Blocks are presented in the CRISMA Catalogue (Havlik et al (Eds.) 2015). Some central BBs are presented below.

Scenario analysis and comparison
The scenario analysis and comparison view consists of several widgets and visually represents the data on indicators and criteria for side-by-side comparison of different simulated scenarios. The vector of indicators is mainly based on selected quantities (e.g., the number of casualties) calculated from a scenario. To be effectively used in the context of decision-support, indicators need to be qualified by assigning to the indicator data the levels of satisfaction. The resulting “normalised” indicators can be better used as decision criteria. As the data on indicators and the data on criteria have the same format (vector of scalar values), both can be displayed in the same way. The following widgets are available for the analysis of a scenario or alternative scenarios if applicable:
• Indicators` and criteria table widget visualises the data on indicators and/or criteria in a table-like view (Figure 12)
• Indicators` and criteria chart widgets allow users to correlate the values of individual indicators and criteria
• Criteria functions` definition widget allows the definition of functions to map indicator values to criteria (Figure 13)
• Criteria spider chart widget shows the data as a spider chart (a.k.a. radar chart) in order to support a quick assessment of the overall performance of the selected scenarios.



Figure 12. Example of indicators-criteria table (Taveter (Ed.) 2015b)



Figure 13. Example of defining a criterion function (Taveter (Ed.) 2015b)

For detailed description and application examples, see the public deliverable D44.3 (Taveter (Ed.) 2015a) and
https://crisma-cat.ait.ac.at/bb/Scenario-Analysis-and-Comparison-View

Multi criteria analysis and decision-support
While the scenario analysis and comparison view allows for the comparison of indicators and criteria for different scenarios, the multi criteria analysis and decision-support view allows for the ranking of different scenarios with respect to a specific decision-making strategy. When doing so, this view introduces complementary decision-support functionalities. The view is composed of the following two widgets:
• The decision strategy widget that allows for defining weighting (emphasis) strategies for different criteria. This way, a weighting factor can be assigned to each indicator (14);
• The decision ranking widget that is used to visualise the actual ranking on the basis of the currently selected criteria functions and decision-making strategy (Figure 15).




Figure 14. Example of decision strategy definition



Figure 15. Example of decision ranking

For detailed description and application examples, see the public deliverable D44.3 (Taveter (Ed.) 2015a) and
https://crisma-cat.ait.ac.at/bb/Multi-Criteria-Analysis-and-Decision-Support-View


3.5 Reference applications for the support of future users

In general, simulation applications built around the CRISMA Framework can support numerous types of decisions. Since CRISMA consists of several components offered by the CRISMA Framework, users may choose, according to their decision-support needs, which BBs they want to include in any particular CRISMA application. Thus, the applications are not only different from the perspective of application architecture but also from the decision-support point of view.

In CRISMA, five reference applications have been set up as templates for the development of crisis management simulation applications for specific crisis domains. The decisions supported in these five CRISMA reference applications reflect the nature and scope as well as particular decision-making needs by the respective end users. They were also used in CRISMA for experimentation and testing in order to support decision-makers in:
• Exploring the impacts of natural disasters in a multi-risk framework including cascading effects;
• Analysing different strategies and mitigation options as the result of crisis management decisions.

The reference applications can be divided into the categories of long-term planning, short-term resource management planning, and training. Table 1 provides an overview of the simulation scenarios considered in the CRISMA reference applications. Various aspects of the decision-support approach applied in CRISMA are discussed in e.g. D44.2 (Engelbach (Ed.) 2014), D44.3 (Taveter (Ed.) 2015a), Heikkilä et al (Eds. 2015), Havlik, et al (2015), Dihé, et al (2013), Engelbach, et al (2014), Erlich, et al (2015), Garcia-Aristizabal, et al (2015), Rosqvist, et al (2015), and Honkavuo, et al (2015).

Table 1. CRISMA reference applications and implemented simulation scenarios


As an example, the following types of questions can be answered through executing different reference applications built around the CRISMA Framework (Heikkilä et al (Eds.), 2015):
• In the Nordic winter storm reference application:
o Is there a need for evacuation? How to time it? Which areas should be prioritised in evacuation?
o What would be the necessary or optimal level of resources for the evacuation of a given area?
o What would be the best response strategy out of the predefined preparedness plans?
• In the Coastal submersion reference application:
o What are the areas with the highest risk of flooding?
o What would be the effect of various long-term measures (e.g., improving dykes) on the flooding risk?
o How effective is the evacuation and temporary sheltering of the endangered population?
o What would be the best long-term planning and investment strategy?
• In the Accidental pollution reference application for the decision-support of mission commanders:
o What would be the optimal way to use the limited number of emergency and rescue resources available in the area?
o How much time is required for different operations? How much time is needed for emergency and rescue resources to arrive at the scene?
o How to take into account that the states of the simulated victims change over time? How to make sure that the most badly poisoned victims get timely treatment?
• In the Earthquake and forest fire reference application:
o What would be the expected damage on buildings due to a given earthquake?
o What would be the expected number of casualties due to a given earthquake?
o What would be the probabilities and effects of cascading events’ scenarios such as a forest fire?
o What would be the effects of different mitigation strategies?
• In the Resource management training and resource planning reference application:
o What would be the probable outcome of the “mass casualty incident” with respect to the accident severity, geographic location and the tactics chosen by the user?
o Which resources should be allocated to which tasks to reach the best possible outcome?
o In addition, the application enables the recording and assessment of first responders’ activities during an exercise.

These five reference applications have been created to demonstrate how versatile and flexible CRISMA Framework is. They also act as examples for any other applications that could be built with CRISMA for end-user needs.


4 EXPERIMENTATION AND TESTING

The testing and experimentation aimed at implementing and evaluating the CRISMA Framework in real-life pilot cases. The testing and experimentation was conducted in two iteration rounds, between which the framework and its technicalities were updated and further developed based on the end-user feedback.

The CRISMA Framework was tested in five very different and complex pilot cases that combined different selections of CRISMA models and BBs with real-life legacy systems. The used scenarios were looking for answers and advice to questions presented in Chapter 3.5 above. For the pilot applications, the CRISMA reference applications were utilised as presented in Figure 16. In pilot demonstrations on both rounds, external end-users have been involved and their feedback has a valuable asset in the further development of CRISMA results.



Figure 16. How CRISMA Pilots have utilised CRISMA Reference Applications

As a result of the testing and validation in five different crisis domains, the final conclusion was that different crisis management applications built around the CRISMA Framework can really support numerous types of decisions. In general, decision-support functionalities of the CRISMA Framework enable the end-users to analyse and compare different simulation scenarios in order to identify sound and efficient mitigation strategies. These mitigation strategies can either be included in preparedness plans or used to e.g. train decision-makers and other stakeholders. In particular, it was proven that the five reference applications helped decision-makers:
• By allowing experimentation with different crisis management strategies and decisions;
• By supporting decision-makers in long-term planning;
• By enabling simulation of natural disasters in a multi-risk framework including cascading effects;
• By facilitating resource management and planning.

It was greatly appreciated that CRISMA decision-support and the related software emphasised the fact that criteria and ranking functions represent opinions rather than facts. In the real life, both the criteria and ranking functions are highly situation-dependent and different stakeholders are likely to disagree on definitions and relative importance of different criteria. The approach chosen in CRISMA encouraged the end users to define several sets of criteria and ranking functions and compare the outcomes of applying them in different alternative simulation scenarios. This functionality in the generic CRISMA Framework is based on the methodology for simulation-based decision support.


REFERENCES

Almeida, M. & Viegas, D (Eds.) 2014. Database and Model for Dynamic scenario assessment V2. Deliverable D42.3 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D423_public.pdf

Cabal, A. (Ed.) 2012. Existing hazard and vulnerability / losses models. Deliverable D41.1 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D411_public.pdf

Dihe, P. (Ed.). 2015. ICMS Framework V2. Deliverable D35.2 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D352_public.pdf.

Dihé, P., Denzer, R., Polese, M., Heikkilä, A.-M, Havlik, D., Sautter, J., Hell, Th., Schlobinski, S., Zuccaro, G., Engelbach, W. 2013. An architecture for integrated crisis management simulation. In: Proceedings of the 20th International Congress on Modelling and Simulation (MODSIM2013), Adelaide, Australia (pp. 873-879).

Dujmović, J. 2007. Properties of Local Andness/Orness. Theoretical Advances and Applications of Fuzzy Logic and Soft Computing Advances in Soft Computing Volume 42, 2007, pp 54-63.

Engelbach, W. (Ed.) 2014. Decision Support and Simulation Model. Deliverable D44.2 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D442_public.pdf

Engelbach, W., Frings, S., Molarius, R., Aubrecht, C., Meriste, M. & Perrels, A. 2014. Indicators to compare simulated crisis management strategies. In: M. Stal, D. Sigrist, W. Ammann (Eds.). Proceedings of the 5th International Disaster and Risk Conference: Integrative Risk Management – The Role of Science, Technology and Practice, IDRC Davos, Davos, Switzerland, 24-28 August (225–228). Global Risk Forum (GRF).

Erlich, M., Cabal, A., Coulet, C., Daou, M-P., Grisel, M., de Groof, A., Havlik, D., Aubrecht, C., Steinnocher, K., Schlobinski, S. & Zuccaro, G. 2015. CRISMA Framework as an experimental tool for simulation of coastal submersion impacts and preparedness for crisis management. In: E-proceedings of the 36th IAHR World Congress, 28 June – 3 July, The Hague, The Netherlands.

Garcia-Aristizabal, A., Almeida, M. Polese, M., Zuccaro, G. & Aubrecht, C. 2015. Improving emergency preparedness with simulation of cascading events scenarios, In: Proceedings of the ISCRAM Conference, May 24–27, 2015, Kristiansand, Norway.

Havlik, D., Deri, O., Rannat, K., Warum, M., Rafalowski, C., Taveter, K., Kutschera, P. & Meriste, M. 2015. Training Support for Crisis Managers with Elements of Serious Gaming. In: R. Denzer, R. M. Argent, G. Schimak, J. Hřebíček (Eds.). Environmental Software Systems. Infrastructures, Services and Applications (217–225). Springer.

Havlik, D., Dihé, P., Frings, S., Steinnocher, K., Aubrecht, C. (Eds.) 2015. Catalogue of CRISMA Applications, Framework Building Block Specifications and Software Implementations. CRISMA consortium. https://crisma-cat.ait.ac.at/content/crisma-catalogue-book.

Heikkilä, A.-M., Havlik, D. & Schlobinski, S. (Eds.). 2015. Modelling crisis management for improved action and preparedness, VTT Technology 228. Espoo, Finland: VTT Technical Research Centre of Finland Ltd. http://www.vtt.fi/inf/pdf/technology/2015/T228.pdf

Honkavuo, H., Jähi, M., Kosonen, A., Piira, K., Rannat, K., Soininen, J., Meriste, M. & Taveter, K. 2015. Enhancing the quality of contingency planning by simulation. In: Proceedings of the ISCRAM Conference, May 24–27, 2015, Kristiansand, Norway.

Polese, M. & Zuccaro, G. (Eds.) 2013. Version 1 of Dynamic vulnerability functions, Systemic vulnerability, and Social vulnerability. Deliverable D43.1 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D431_public.pdf

Polese, M. & Zuccaro, G (Eds.). 2014. Version 2 of Dynamic vulnerability functions, Systemic vulnerability, and Social vulnerability. Deliverable D43.2 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D432_public.pdf

Rosqvist T., Meriste M. & Havlik D. 2015. Reference decision model for simulation of situational awareness related to first response. In: Proceedings of the International Forum for the Military Training, Education and Simulation Sectors (ITEC), Czech Republic, Prague, 28–30 April.

Taveter, K. (Ed.) 2015a. Decision Support and Simulation Model. Deliverable D44.3 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D443_public.pdf

Taveter, K. (Ed.) 2015b. Validated simulation tool for crisis management strategies and planned actions. Deliverable D44.4 of the European Integrated Project CRISMA, FP7-SECURITY-284552. http://www.crismaproject.eu/deliverables/CRISMA_D444_public.pdf

Yager, R.R. 1988. On ordered weighted averaging aggregation operators in multi-criteria decision-making. IEEE Transactions on Systems, Man and Cybernetics, 18 (1), 183–190.

Yager, R.R. 1996. Quantifier guided aggregation using OWA operators. International Journal of Intelligent Systems, pp. 11‒49.

Zuccaro, G. & Filomena, P. 1997. Multi-criteria analysis in vulnerability assessment. Second Egyptian Conference on Earthquake Engineering, Aswan, Egypt

Potential Impact:
In recent years disasters have significantly affected populations and their living environments globally. Decision makers at different levels appeared to be unprepared for responding adequately to new magnitudes of crisis management demands. Natural and human-made disasters are not events, but processes, in which previous historical responses to events contribute heavily to the degree of preparedness and the extent and nature of impacts and the recovery from disasters offers important opportunities to address underlying causes, problems and institutional incapacities. The extent of the impact of a disaster is closely related to the capacity of institutions and the public to learn and adjust from previous experiences, and on the pre-disaster performance and capacity of infrastructure – transport systems, electricity and water supply. The more that policy makers and citizens understand the crisis evolvement and likely impacts, the more they will prepare to increasing risks. It is apparent that policy level decision making plays a large role in the degree of preparedness, and that the complexity of the situation and impacts call for multi-sectoral coordination between authorities and other stakeholders.

Developing capacity in emergencies and reconstruction settings is the key to mitigating the otherwise disastrous consequences of low capacity and preparedness. Crisis simulation (planning and training) offered by CRISMA helps to justify the necessary improvements in the pre-disaster condition and by creating opportunity to learn and understand the crisis situation, which lead to improved readiness for quick decision making and quicker help.

In the management of large scale crisis events, there is a demand for increasing efficiency and effectiveness of the overall decision making processes. As an open decision support and simulation environment the CRISMA will provide what if scenario support to evaluate the consequences of mitigating actions and to enable collaborative assessment of complex situations and provide the means to publish and share crisis scenarios with other users involved in the decision making process across organisations and national borders. CRISMA will provide an Open Crisis Management System that will be able to integrate and make use of existing crisis management software components to be able to assess complex and multi-sectoral effects of a crisis.

In decision-making related to complex crises, decisions are made not on the basis of a single event, but rather correlating multiple events into a complex event and mapping it to a situation happening in the operational space. The CRISMA system allows an easy start for crisis management simulation (planning and training) by providing generic rules and assumptions as well as reference crisis scenarios based on generalised hazard models that can be easily applied to other areas. On that basis, the organisations also have a baseline to decide on investments in models specific to their region and their own thread scenarios. The system allows improved simulation and result visualisation of alternative options for staff and technology resources. This encourages the strategic and political discussion on adequate civil protection capacities and provides a scientific background for general alternatives.

Crisis management organisations will be able to use their own already gathered and existing data for the purposes of planning, training and operation by easily integrating them with the help of the CRISMA Framework and its simulation and modelling facilities; CRISMA on the other hand provides a set of standard criteria and reference figures that allow an easy start without in depth customising at the very beginning.

End-User Benefits
Through CRISMA, end-users benefit the possibility to better understand the evolvement of a crisis and the mitigating effects of their possible decisions and actions. End-Users will be supported in collaborative decision making which is a key in the handling of large crises where a multitude of experts, organisations, and public authorities is involved. By using CRISMA End-Users will gain access to a wide variety of information sources which is necessary to support ad hoc decisions (not planned events). This leads to a better understanding of the available resources and the interrelations between countermeasures by the involved crisis managers from different organisations and thus to a better cooperation in real crisis situations. CRISMA enables the crisis management to argue for the potential benefits of preparedness and better resources, since key performance indicators and cost benefit analysis of alternative strategies to manage the identical crisis scenario are provided.

Socio-economic benefits
Socio-economic benefits of CRISMA are eventually gained by the citizens, through better preparedness, response and contingency plans. Preventative actions and awareness of possible hazardous events will improve the resilience of a society and its individual members. The risk profiles of increasingly large and dense urban centres indicate that the vulnerability of urban population can be enormous, as demonstrated by the large numbers of victims of earthquake events in China, Pakistan and Haiti. The understanding and awareness on this pattern and regional vulnerability offers a starting point to the improved preparedness planning and risk-informed decision making of CRISMA Tool.

CRISMA provides an efficient framework to support training of personnel in charge of managing complex crisis. They can actually be exposed to a wide set of situations based on real cases or simulated ones to assess the degree of preparedness, the effect of decisions taken on the development of the situation etc. An exercise rarely done by governments but extremely useful is to conduct a disaster simulation with the various parties involved in post disaster financing and assistance. Such simulation invariably helps identify bottlenecks and weaknesses in existing budget processes, emergency procurement, contract monitoring, and payment systems, among other aspects. It also helps sensitise public officials, particularly in finance ministries, who are rarely confronted with disaster emergencies, and helps improve preparedness at all levels of the government.

Economic benefits
Crisis management is a key sector under the overall security business activities, and the market is growing along emerging threats and expansive consequences of hazards. Hardly any mission could succeed without private companies and contractors.

Services delivered by private companies include medical care, logistics, transportation and information systems and infrastructure (Kupi 2010). Wide spectrum of existing national information systems and command and control tools require high flexibility and open integration interfaces. As the threats are changing, decision support systems should include a provision not only for existing but also future decision-making situations. It also important to look the whole continuum and identify flexible business models, solutions and technologies for both the crisis management operations, preparedness planning, as well as for the rebuilding process. In the area of crisis management private companies cooperate with local authorities, governments, international organisations and civil-society actors. Public-private partnerships therefore play an important role in the crisis management business and also in building up preparedness. Information technology based solutions, which deliver shared situational awareness and open interfaces for integration, and thus support timely decision making and the use of GUI information, are relevant though the whole crisis cycle from prevention and preparedness beyond the acute rescue actions. While the CRISMA project focuses on planning for preparedness and response, the results benefit other steps of the crisis cycle too.

Dissemination and exploitation of CRISMA

During the project, the acceptance and further exploitation potential of CRISMA was aimed to be ensured through constant dialogue and contacts with external end-user and other stakeholder communities. CRISMA has organised external end-user workshops as well as trainings for both potential end-users and technology providers in crisis management sectors. In these events end-user communities have guided CRISMA developments through feedback on both basic assumptions and chosen approaches of the Consortium. Also the five pilot demonstrations at the end of the project have been open for non-project stakeholders to evaluate how CRISMA results meet needs on the field.

The CRISMA Consortium and the representative end-users involved in the pilots believe that CRISMA has delivered a valuable contribution to provide multi-sectoral anticipation of the possible consequences of an incident and capabilities to simulate the results of various alternative measures and actions. Due to the scope and significance of the challenges addressed, the principal target users are actually the decision-makers at national and European level. It can be seen that the market for crisis management related planning and command and control room applications will be under change in coming years. The CRISMA system will enhance European understanding of the new requirements and build some of the needed solutions and therefore enhance possibilities for European companies in succeeding in this challenging segment.

Currently the CRISMA industrial partners are supporting crisis managers as well as rescue and security forces with a broad range of equipment to improve their operational abilities and efficiency and to reduce their response time. Still there is a gap between crisis management, dedicated and organised reaction on crisis situations and the appropriate command and control of local rescue and security forces. The CRISMA project will allow to close this gap by providing an environment that allows examination, management and planning of operations for crisis situation in detail. More specifically:
• To facilitate the monitoring of the planned operations, the integrated visualisation tools allow to monitor planning of operations and to evaluate appearance of certain crisis typical situations. CRISMA will allow adapting the tool with respect to crisis scenarios.
• To facilitate the identification of capability gaps, CRISMA will identify capability gaps that can be transferred to operational needs for rescue and security forces, especially in the domain of command and control. This can be used by the CRISMA industrial partners as well as the various end-users to improve overall processes and equipment required to act on future crisis situations in an appropriate way.

CRISMA offer is built upon the innovative concepts and features covering methodological and technological aspects of the crisis management approach. CRISMA innovations – CRISMA Framework toolkit, integrated solutions, individual SW components, and expert services – bring valuable advantages to the end-users. The wider exploitation of CRISMA and further development of the CRISMA Framework to meet future challenges is facilitated through its framework approach that allows other model and tool developers to enter the marketing via the CRISMA solution. The design and building blocks of the Framework are mainly available as open source and supported by CRISMA industrial partners, constituting the core of a developers’ community.

List of Websites:
http://www.crismaproject.eu/

Contact

Heikkilä, Anna-Mari (Senior Research Scientist)
Tel.: +358 20 7223490
E-mail

Subjects

Security
Record Number: 186860 / Last updated on: 2016-07-13