As you support a business-critical application deployed on Cloud Run in production, the application reports HTTP 500 errors, impacting its usability. You aim to receive alerts when the error rate exceeds 15% of the requests within a specific time frame. What should be your approach?