Designing a Hybrid Optimization Algorithm for Improving QoS in Flood Prediction Using Cloud Computing

The field of natural and environmental sciences garners significant attention due to its reliance on precise, real-time predictions. Among the various environmental hazards, flood events caused by intense rainfall represent a recurring threat, particularly in the eastern regions of India. This study focuses on where frequent flooding has affected numerous districts, resulting in substantial loss of life and damage to property. Cloud computing is a technology that enables the provision of several types of computer services over the internet. They are often maintained by a cloud services provider (CSP) at a remote data center for maintaining the flood based data. While cloud computing offers numerous advantages, there are also some disadvantages and challenges associated with services delivered to cloud users affecting the QoS. Integrating services from multiple vendors can be challenging because each cloud provider has its own set of services andit may be a challenge for some organizations to tailor cloud services to their specific business demands.  To overcome such issues related to service delivery, a Multi agent-based hybrid algorithm combining both ACO and PSO has been proposed which enhances the QoS by finding an optimum execution time. A cloud platform for flood prediction has been created in which agents interact with each other to optimally cater to the user's requests. An ablation study is done with flood data signals for measuring multi-agent functionality in cloud using the proposed Hybrid ACO and PSO algorithms. Experimental results shows effective execution of user tasks using agent with an improvement of 50% than that of existing algorithms.