Limits the harm caused by incidents (unplanned interruptions to a service, or drops in its quality) by getting normal service back as fast as it can. Each incident is logged and handled against agreed target resolution times, prioritized according to business impact and an agreed classification, with good updates. Resolution may come through self-help, the service desk, support teams to which it is escalated and routed by category, suppliers and partners, temporary teams for the most complex incidents and for every major one, or, in extreme cases, invoking disaster recovery, with swarming and collaboration tools used where they help. Incident records feed improvement according to frequency and severity; incidents are visible to users and customers, so communication sets expectations and confirms closure; incidents in test, release and development environments are resolved under control; and resolving incidents is a large part of deliver and support.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.