A highway viaduct is a double-line reinforced concrete structure,and the bridge deck width of a single-line is 12 m.The northern section and the southern section of the bridge approach and the main bridge have 6 piers(pier number 4-9),5 piers(pier number 12-16) and 2 piers(pier number 10,11),respectively.When blasting demolition of this bridge is conducted,it is necessary to focus on the complex surrounding environment,flying rocks and touchdown ground vibrations.The finite element software ANSYS/LS-DYNA is used to analyze the combined effect of the blasting load and the intrusion load of the fragmentation group on the reinforced concrete pier columns of the main bridge,and simulate the whole collapse process of the bridge.As a key point of the study,three blasting schemes,symmetrical initiation,unidirectional initiation,and center initiation,are compared,and the muck pile parameters are evaluated by using the post-processing software LS-Prepost.It is revealed that if only consider the muck pile,center initiation performs the best.Between the other two blasting schemes,unidirectional initiation is better than symmetrical initiation.The peak particle velocities of touchdown ground vibrations for the three blasting schemes can be ranked in an descending order by numerical simulation,which is symmetrical initiation>center initiation>unidirectional initiation.At the same time,all the PPVs of the three blasting schemes are within the safety allowable ranges.After comprehensively evaluating the muck pile parameters and touchdown ground vibrations,it is determined that a long delay interval has a better blasting effect than a short delay interval.Finally,the blasting scheme of unidirectional initiation,long delay time within and between the bridge lines has been chosen.In general,the actual blasting effect coincides with the numerical simulation results,with all indicators in line with the corresponding regulations.In addition,a good vibration reduction effect is also achieved by using various vibration damping techniques.