Exploring Cloud-Based Distributed Disaster Management With Dynamic Multi-Agents Workflow System

Abstract

Natural disaster is one of the important topics in current researches. Disaster Management System (DMS) is a complex system and needs to perform a collection of tasks collaboratively along with the potentiality to change the configurations of the system dynamically. In the research era of workflow model, existing models mainly deal with temporal and static constrains. However they cannot be used to keep pace with an uncertainly dynamic system like disaster management. Considering all these significant DMS attributes we have designed a new dynamically configurable and changeable workflow model with the support of adaptive scheduling, for both successful and failed situations, and implemented in a distributed cloud system to maintain the rescue and reorganization activities of disaster situation. In order to simplify the system architecture, we have used Multi Agent System (MAS) for our design. The proposed system achieves a comparatively higher rate of successful job completion-higher rescheduling success rate and comparatively lower dropout rate.