A comprehensive framework (from real-time prognostics to maintenance decisions) studying the influence of the imperfect prognostics information on maintenance decision is an underexplored area. Thus, we bridge the gap and propose a new comprehensive maintenance support system. First, a new sensor-based prognostics module was modelled employing the Weibull time-to-event recurrent neural network. In which, the prognostics competence was enhanced by predicting the parameters of failure distribution despite a single time-to-failure. In conjunction, new predictive maintenance (PdM) planning model was framed through a tradeoff between corrective maintenance and lost remaining life due to PdM. This optimises the time for maintenance via all gathered operational and maintenance cost parameters from the historical data. Its performance is highlighted with a case study on maintenance planning of cutting tools within a manufacturing facility. We provide systematic sensitivity analysis and discuss the impact of the imperfect prognostics information on maintenance decisions. Results show that uncertainty, regarding prediction, drops as time goes on; and as the uncertainty drops, the maintenance timing gets closer to the remaining useful life. This is expected as the risk of making the wrong decision decreases.