This is a cost-value calculation and neither approach is "best" at all cases. Does the cost of developing the tools and procedures exceed the cost of planned do

This is a cost-value calculation and neither approach is "best" at all cases. Does the cost of developing the tools and procedures exceed the cost of planned downtimes during the next X years? Often you build some tools that mitigate some of the risk and plan deployments at a less active hour to mitigate the rest of it.