Your application is deployed on Google Cloud Platform (GCP) within a Managed Instance Group (MIG). Recently, you have observed specific error messages in Cloud Logging indicating that one of the virtual machines (VMs) in the MIG has a non-responsive process. To swiftly address this issue, you need to replace the problematic VM within the MIG. What steps should you follow to achieve this?