analyzing historical data for one or more applications on an ongoing basis, using a machine learning model, to identify one or more parameters affecting system quality, and/or one or more parameters affecting system performance, and/or one or more parameters affecting system stability;

generating, for each of the one or more applications, a change risk score for each application based at least in part on a weighted average of one or more parameters affecting system quality, a weighted average of one or more parameters affecting system performance, and a weighted average of one or more parameters affecting system stability;

categorizing a risk for each of the one or more applications based upon the change risk score;

determining for each of the one or more applications whether changes for an application of the one or more applications can be deployed based on the categorized risk of the application's change risk score; such that

if changes for the application can be deployed, performing an automated deployment process that deploys the changes for the application; and

if changes for the application cannot be deployed, preventing the deployment of any changes for the application;

using new historical data, as the new historical data becomes available, on an ongoing basis, to update the machine learning model, so as to allow the analyzing to more accurately identify updated parameters; and

using the machine learning model to perform pattern recognition to recognize patterns of incidents and their root causes over time to identify common deployment related issues and their contributing parameters.