Online remaining useful life (RUL) is a major challenge of prognostic and health management systems (PHM) in many industrial domains where safety, reliability and cost reduction are of high importance. To reduce the cost, one solution is to match the maintenance date with the estimated remaining life of the system. This prediction of the RUL allows fixing time in the future to organize a maintenance action which can be called maintenance time-window. Nevertheless, the RUL can change due to the different dynamics of the operating conditions over the operation time. It may shift the maintenance window into another time. System health should be updated when new measurements come in analysis and the prognostic shows an updated RUL. Thus, the online RUL prediction is a much more effective approach in condition-based maintenance. This paper presents an Input-Output Hidden Markov Model (IOHMM) that estimates the online prognostic based on passed to current measured data from the system which are used to manage the RUL that corresponds to a target remaining time. A reference manager is designed to figure out the next input condition according to the new measurements in order to reschedule the maintenance time window. An example is shown in which well-known algorithms dedicated to HMM are adapted to IOHMM for online prognostic when the system emits a new observation.