Fault tolerant methods using dynamically reconfigurable devices have been studied to overcome wear-out failures. However, quantitative comparisons have not been sufficiently assessed on device lifetime enhancement with these methods, whereas they have mainly been evaluated individually from various viewpoints such as additional hardware overheads, performance, and downtime for fault recovery. This paper presents quantitative lifetime evaluations performed by simulating the fault-avoidance procedures of five representative methods under the same conditions in wear-out scenarios, applications, and device architecture. The simulation results indicated that improvements of up to 70% mean-time-to-failure (MTTF) in comparison with ideal fault avoidance could be achieved by using methods of fault avoidance with 'row direction shift' and 'dynamic partial reconfiguration'. 'Column shift', on the other hand, attained a high degree of stability with moderate improvements in MTTF. The experimental results also revealed that spare basic elements (BEs) should be prevented from aging so that improvements in MTTF would not be adversely affected. Moreover, we found that the selection of initial mappings guided by wire utilization could increase the lifetimes of partial reconfiguration based fault avoidance.