Wyniki wyszukiwania - BazTech

1

Decision Making in Crisis Management : Time Criticality Assessment

Eid M., Souza de Cursi E., Abdelkhalak E.

Journal of Polish Safety and Reliability Association

|

2017

|

Vol. 8, No. 1

13--22

EN

The paper is proposing a methodology for assessing the time-criticality in the scope of a crisis resulting from a cascade of disruptions of a set of Critical Infrastructures (CIs). The success of the management of the crisis depends on taking the good decisions and undertaking the corresponding actions at the good timing. Identifying time intervals for decisions and actions requires dynamic models capable to assess the crisis time evolution. While, measuring the “criticality” of different time-intervals requires the use of suitable metrics. Based on previous works for modelling cascade of CIs’ disruptions incorporating CIs dependency and vulnerability, the authors propose a methodology to assess the time-criticality and propose the appropriate metrics. The methodology is based on the idea of comparing the time-profile of a given cascade of disruptions between two configurations: unstressed and stressed CIs. The CIs become stressed under the action of a threat and combined with the dependency between CIs. The unstressed configuration represents a risk-noise. Two metrics are proposed in order to carry on the comparison between the time-profiles of the stressed and the unstressed CIs. The proposed metrics are: the time to attend 90% of the asymptotic occurrence probability and the time to attend the most probable occurrence rate, describing the cascade likelihood. An academic case is presented in order to demonstrate the applicability of the methodology and illustrate some interesting features.

2

Critical Infrastructures Protection (CIP) : coupled modelling for threats and resilience

Eid M., El-Hami A., Souza de Cursi E., Kołowrocki K., Kuligowska E., Soszyńska-Budny J.

Journal of Polish Safety and Reliability Association

|

2015

|

Vol. 6, No. 1

85--94

EN

Critical Infrastructure Protection (CIP) requires the development of models and tools able to describe and simulate the threats’ and the Critical Infrastructure’s (CI) dynamic behaviour. However, these modelling activities are very often carried out separately for threats and for CI behaviour. An effective assessment of the CI’s resilience and preparedness requires real coupled models of threats dynamic and CI’s one. The authors develop some basic ideas about coupled modelling in the sense of coupling the dynamics of the threat with that of the CIs, within a stochastic modelling approach. Such coupled dynamic models would enhance the effectiveness of our capabilities to assess CI’s resilience and to help in decision making for crisis management.

3

A topological model to assess networks connectivity and reliability

Eid M., Hami-EL A., Souza de Cursi E.

Journal of Polish Safety and Reliability Association

|

2013

|

Vol. 4, No. 2

167--178

EN

The acceleration of the interconnectivity of networks of all sorts brings to the front scene the issue of networks performance measure. Recently, one observed an accelerating course towards quantitative probabilistic models to describe and assess networks’ Connectivity, as being the main vector of performance. However, modelling realistic networks is still far from being satisfactorily achieved using quantitative probabilistic models. On the other hand, little room had been lift to exploring the potential of topological models to develop qualitative and semi- quantitative models in order to assess networks connectivity. In this paper, we are exploring the potential of the topological modelling. The proposed model is based on describing the nodepair connectivity using binary scalars of different orders (tensors). Preliminary results of our explorations sounded very promoting.

4

Random perturbation of the projected variable metric method for nonsmooth nonconvex optimization problems with linear constraints

El Mouatasim A., Ellaia R., Souza de Cursi E.

International Journal of Applied Mathematics and Computer Science

|

2011

|

Vol. 21, no 2

317-329

EN

We present a random perturbation of the projected variable metric method for solving linearly constrained nonsmooth (i.e., nondifferentiable) nonconvex optimization problems, and we establish the convergence to a global minimum for a locally Lipschitz continuous objective function which may be nondifferentiable on a countable set of points. Numerical results show the effectiveness of the proposed approach.