Restoration of pine forests has become a priority for managers who are beginning to embrace ideas of highly heterogeneous forest structures that potentially encourages high levels of regeneration. This study utilizes stem-mapped stands to assess how simulated regeneration timing and magnitude influence longevity of reduced fire behavior by linking growth and yield model outputs to a crown fire prediction model. Treatment longevity was assessed as return time to within 10% of pre-treatment predicted wind speeds for the onset of passive (Torching) and active (Crowning) crown fire behavior. Treatment longevity in terms of Torching and Crowning was reduced 5 years for every 550 and 150 seedlings ha-1, respectively. Introducing regeneration as a single pulse further reduced Torching treatment longevity 10 years compared to other regeneration distributions. Crowning treatment longevity increased at higher site indices, where a 6 m increase in site index increased longevity 4.5 year. This result was contrary to expectations that canopy openings after treatments would close faster on higher productivity sites. Additionally, Torching longevity was influenced by the rate of crown recession, were reducing the recession rate decreased longevity in areas with higher site indices. These dependencies highlight a need for research exploring stand development in heterogeneous sites.