To mitigate the level of blast-induced vibrations, the time-delay experiments involved 5 delay intervals, 8 holes and 11 monitoring stations, were undertaken in a newly constructed underground mine. The method of instantaneous energy based on the empirical mode decomposition (EMD) was adopted to identify the actual delay times from the near-field accelerations. Then the Theil-Sen regression was adopted to analyze the vibration attenuation. Based on the drawbacks of the field experiments due to various unpredictable factors, the blast damage model was further employed to make a comprehensive understanding of the effects of delay times. The signature invoked the average waveform of multiple holes in the experimental 75 ms case, then a group of fluctuating signature waveforms were generated from the Monte Carlo scheme based on waveform comparison between the measured and the signature. Upon completion of the model verification, the number of delays was examined within four delay intervals, and 25 delays were selected to figure out the effects of delay times and delay errors. The results state that the peak particle velocity (PPV) and average frequency (AF) are insensitive to delay times irrespective of delay errors, except short delay times. The vibration dominant frequency (DF) is mainly contributed by idea delay times and delay errors, and the optimal delay interval is half of the dominant period, which is only effective to improve DFs at a target location. The commonly used pyrotechnic detonator is competent in most blasts unless structures need to be protected from their resonances.