unavailability

Unavailability is the state of a system, service, or resource being inaccessible or non-operational when needed. In reliability engineering and operations management, unavailability is used to quantify the fraction of time a system is not available to perform its intended function within a given period.

It is commonly expressed as a percentage or rate and is the complement of availability. When availability is defined as MTBF divided by (MTBF plus MTTR), unavailability can be calculated as MTTR divided by (MTBF plus MTTR).

Unavailability comprises planned downtime for maintenance or upgrades and unplanned downtime caused by failures, faults, or outages. Planned downtime is often scheduled and announced under service level agreements; unplanned downtime is typically mitigated through redundancy, monitoring, and failover mechanisms.

Contexts: Information technology, telecommunications, manufacturing, and service industries track unavailability to assess reliability, risk, and user impact. In cloud services, downtime is a critical SLA metric; in manufacturing, equipment unavailability limits capacity.

Mitigation strategies include redundancy, fault tolerance, predictive maintenance, proactive monitoring, automated recovery, and disaster recovery planning. Understanding and reducing unavailability supports reliability goals, customer satisfaction, and operational performance.