LAUSR

The main purpose of this paper is to design a scientific-based probabilistic model based on Markov chains, to calculate reliability indicators, such as mean time between failure (MTBF) and mean time to failure (MTTF) based on probabilities, to compute failure rates based on statistical data, and to provide an algorithm to calculate the maximum number of interruptions and the maximum duration of one interruption for a photovoltaic power plant (PV-PP) meant to improve the operation and maintenance (O&M) activities and optimize the stocks of the spare parts. Over almost two years, events are recorded at PV-PP Agigea with an installed power of 0.5 MW. The predictive maintenance of future events and stockpile sizing at the PV-PP Agigea are developed taking into account the maintenance activities carried out at the plant’s components since the commissioning of the PV-PP Agigea. Based on the PV-PP data and the intervention reports that consist of the incidents recorded between February 2016 and December 2017, it is intended to determine by statistical methods the following basic reliability indicators: failure rate, usually symbolized by $\lambda $ , defined as the average number of failures on time unit, and the maximum number of interruptions ( ${N} _{{max}}$ ) eliminated through repairs or replacements during the reference period, determined for a certain level of risk.