Implementing self-healing N-policy queueing models and their impact on IoT design applications

Abstract

We embarked on a comprehensive exploration of a single server queueing design, with a specific focus on handling soft failures. Soft failures refer to instances where customers do not need to be removed but rather need to wait for the server to be reactivated. These occurrences can happen at any time during the server's operation. When a soft failure occurs, the process automatically initiates a repair action, which we will refer to as the self-healing time. This self-healing time is relatively short, as the server possesses a remarkable restoration capability. Once the repair is complete, the server resumes its service provision and resumes normal operations. Moreover, during periods of prolonged idleness, the server can enter a dormant state, akin to a vacation mode. This dormant state is triggered when the server awaits the accumulation of N or more users. Once the threshold is reached, the server transitions into a busy state and resumes its normal operations. This study represents the pioneering integration of soft failures with the N policy, marking the first of its kind in this field. Additionally, we provide explicit expressions for the transient probabilities of the model, employing generating function methodology and Laplace transform techniques. Furthermore, we include performance measures and a numerical component to underscore the significance of the model's parameters.