A social media company is running its flagship application via an Auto-Scaling group (ASG) which has 15 EC2 instances spanning across 3 Availability Zones (AZs). The current average CPU utilization of the group sits at 15% off-peak time. During peak time, it goes all the way to 45%, and these peak times happen predictably during business hours. The company has hired you as an AWS Certified DevOps Engineer Professional to build a solution for this requirement. How can you improve the instance utilization while reducing cost and maintaining application availability?