The concept of resilience and various aspects related to resilience enhancement and resilience calculus are addressed in the context of the broader theme of reliability and disaster mitigation. A definition of resilience is proposed, based on the assessment of event trees and the statistical determination of risks. Essentially, we apply standard probabilistic models with conditional probabilities, event trees and fault trees, where faults are not limited to technological ones, but cover lack of personnel as well, to define the resilience in a systematic manner. An essential aspect of the approach is the estimation of the probability of recovery in a specified time frame. The time to recovery and the level of recovery play key roles in the definition. The main assumptions underlying the calculations are discussed. A synthetic resilience indicator is introduced to offer a short, easy to grasp clue on the resilience level, with the aim to facilitate planning for resilience. An analysis methodology is suggested under general conditions, and several examples of applications are given.