Sometimes it happens that a PMF run does not constrain. Non-convergent solutions remain empty after having been imported and this is also reported in the log file as well as in the command line.
Most of the time, convergence is not achieved because the error matrix is wrongly estimated or unsuitable PMF settings were chosen. So in a first step, the error matrix should be verified. Also the option of normalization during the PMF also often results in non-convergent runs. Rerunning the model using less restrictive convergence criteria should be only considered as a last resort.