The successive removal of units from a group of systems during phasing out leads to a variable requirement for stock level of spare parts. Repairable units from phased out systems can be reused for the remaining functional systems in the group. Hence, the stock level of spare parts increases and the demand for spare parts decreases. This can lead to excessive stocking of spare parts and high maintenance costs for spare parts. This paper proposes a methodology to determine the optimum time to stop repair of repairable units to minimize the maintenance cost. It uses a Poisson Distribution based on NHPP (Non-Homogenous Poisson Process) to predict the available number of units, while fulfilling the demand for spare parts for the remaining systems. A concept, called Minimal Margin is introduced to formulate the problem and nonlinear programming is proposed to obtain the optimum solution. Finally, a numerical example is presented to demonstrate the approach
Godkänd; 2010; 20100611 (ysko)