The concept of minimal repair is that the repair cannot return the system to "as new" condition but instead returns it to the average condition for a working system of its age. When a system requires $n^{th}$ repair, it is first inspected and the repair cost is estimated. The $n^{th}$ minimal repair is undertaken only if the estimated cost is less than the repair cost limit.
This thesis presents a repair cost limit replacement model considering expected failure time interval under minimal repair, where $n^{th}$ repair cost limit is obtained by multiplying α and the expected time interval between the n-$1^{th}$ failure time and the $n^{th}$ failure time. A weibull failure distribution and a negative exponential repair cost distribution are assumed for analytic tractability. The average cost per unit time for repairs and replacement over an infinite planning horizon is applied as a criterion, and their behaviors are examined, and the optimal α is searched by a computer program.