Community Research and Development Information Service - CORDIS


PREDICT Report Summary

Project ID: 607697
Funded under: FP7-SECURITY
Country: France

Periodic Report Summary 1 - PREDICT (PREparing for the Domino effect in Crisis siTuations)

Project Context and Objectives:
The PREDICT project has started on April 1st 2014 and involves 11 partners. Three end-users are part of the project and are directly involved in the three test cases that will validate the PREDICT Suite.

PREDICT will provide a comprehensive solution for dealing with cascading effects in muti-sectorial crisis situations covering aspects of critical infrastructures. The PREDICT solution will be composed of the following three pillars: methodologies, models and software tools. A generic approach will be setup to prevent or mitigate cascading effects that will be applied in selected cases agreed with end-users. The software requires many tools working in the background and at least two tools interfacing with the crisis management (CM): a decision support system (DSS) and a foresight and prediction tool (FPT). Their integration will increase the awareness and understanding of cascading effects by crisis response organizations, enhances their preparedness and improves their response capability to respond in case of cascading failures.

The new methods and tools developed within the PREDICT project may reduce the negative impact of possible future cascading effects and may improve the planning of civil protection and crisis management operations. The PREDICT results will help lowering losses and damages in various fields, including economic or social safety and security. In order to bring this new quality into the cascading effects and crisis management domain, the proposed project will achieve the six operational and technical objectives.

1. Gather and analyse available domain knowledge
2. Develop a common framework
3. Create models of cascading effects and interdependencies
4. Develop a suite of software tools
5. Validate the solution through running simulations
6. Disseminate project results
Project Results:
The PREDICT project started a year ago, the Work-Package on domain analysis and requirements is finished (WP2) all the other WP started. The results of WP2 are the baseline for understanding cascading effects and resilience in complex systems. In this state-of-art research in major areas related to cascading effect and systems resilience, we focused on threats identification, threats specification, critical infrastructures’ dependency and crisis management assessment.

On the metrics used to describe the “threats”, in view of enhancing systems’ resilience, we reviewed natural threats and man-made threats, excluding threats with adaptability-features. That is the case of: terrorist actions, sabotages and wars. The principal finding is that major threats are assessed considering four notions: magnitude of the threat, intensity of its impact (on a well-defined set of systems), its likelihood and its impact lasting time (mean time to heal).

On crisis management assessment, our research concluded that there are no international standards for crisis management, so best practices play a crucial role. The private sector conducts short-term business continuity management and risk management. It appears that in many European countries, the public sector is better prepared for crisis situations than the private sector. Because of the interdependencies between CI sectors, crisis management at a national level requires collaboration in public-private partnerships. Coordination and information exchange are crucial, and barriers need to be removed or weakened.
A deep analysis of lessons learned from crisis response operations and cascading effects, and the database of past cascading events, to identify the essential elements for identification and probability assessment of cascading effects and develop the main steps of the PREDICT methodology to assess cascading effects. The methodology consists of the following steps:

1. Identify the threats;
2. Identify the CI in the region;
3. Identify the key CI elements;
4. Characterize the vulnerability of the key CI elements to the threat;
5. Assess the first order impact of the threat on the CI elements;
6. Describe the dependencies between the CI elements in the region;
• Describe the required input and output of all key CI elements;
• Distinguish between the different modes of operation;
• Include the temporal and spatial factors;
7. Assess the CI cascading effects.

In this project, end-user workshops have been organized to help defining the features of PREDICT tools. For a DSS, we remind that the quality of the results depends directly on the quality of the inputs and their availability. Within a crisis context, the time is the second hard constrain. These two aspects will be optimized in the PREDICT DSS. For the FPT, the first essential requirement is to effectively process the available data. Thus effective and adequate data mining is considered to be the crucial feature of the FPT. Secondly, time-scale and geographical extension are two important factors. A task will be to optimize the time-scale and geographical extension.

The backbone of the tools is now being developed in WP4 in direct cooperation with WP5 and WP6. It is the design of the system architecture suitable for integration of the tools. The integration of the FPT and DSS with the necessary background tasks is not straightforward but a methodology is put in place. This approach follows the top-down Mission-Concepts-Realisation-Implementation (MCRI) scheme elaborated in the EU FP7 projects DIESIS, CIPRNet and CRISMA.
Potential Impact:
The main results after one year are illustrated in the deliverable submitted. On the technical side, the analysis and requirements of WP3 is over and reported in D2.1, D2.2 and D2.3. Methodologies for the identification and probability assessment of cascading effects have been reviewed and are available in D3.1. A great work on the preparation of the test cases has been performed by the consortium including many end-users. This work is detailed in the three following deliverables: D7.1, D7.2 and D8.1.

The ultimate goal of PREDICT is to help avoiding or mitigating cascading effects in critical infrastructures (CIs) in cases of emergencies and crises. The required fundamental steps to achieve that goal include:
o Identify cascading effects: what is the risk that cascading CI effects during a “local” crisis have an impact on crisis management itself and affect response operations and mitigation actions?
o If possible, “show” the evolution of the cascading effect and its impacts to the crisis managers
o If possible, propose a plan to the crisis manager to avoid or mitigate the cascading effects

To that end, PREDICT will develop and deliver prototypes of tools that support crisis management in that task. The added values of the PREDICT tools and integrated tool suite will be:

o Providing information on possible cascading effects of Critical Infrastructures in cases of emergencies and disasters. Effects include in particular, impacts on CM’s mitigation and response operations. The PREDICT tools will, for instance, generate foresight and predictions of incident evolution and predict effects that are relevant for crisis management and responders. Early availability of such information will enable faster reaction and development of adaptation of contingency plans.
o Facilitating information sharing in (distributed) crisis management teams. One end user (VRZHZ) has a computer-based information sharing system in place (LCMS, reference) that connects all crisis management teams. Other end users do not have such computer-based information sharing systems. In WP6, the existing middleware system DPIF (Dynamic Process Integration Framework) and the DPIF-based information-sharing application DEIN will further evolve into an information-sharing platform for PREDICT end-user SYKE.

Typically, crisis management is performed in four phases: response, recovery, mitigation and preparedness. In principle, added value information on cascad-ing effects for CM decision support can be useful in all phases. The PREDICT consortium and its end users agreed on focusing primarily on the support of the response phase, and setting a second focus on training as a key element of the preparedness phase.

List of Websites:


Dominique Sérafin, (Defence Research Advisor)
Tel.: +33 5 65 10 54 46


Record Number: 182653 / Last updated on: 2016-05-17
Information source: SESAM