Hi ,
Agree with Pavan.
Datacenter switchover is the only option.If there is simultaneous failure happened on all the nodes in the primary site or else an primary datacenter failure then there wouldn't be an enough time for the cluster to recalculate and adjust the quorum.So the only
way is to do an datacenter switchover.
Reference link for datacenter Switchover :
http://smtpport25.wordpress.com/2010/12/10/exchange-2010-dag-local-and-site-drfailover-and-fail-back/
Above mentioned article is for exchange 2010 even though the steps defined on that article is same for exchange 2013.
Reference kink for Dynamic quorum :
http://www.msexchange.org/articles-tutorials/exchange-server-2013/high-availability-recovery/exchange-2013-dag-dynamic-quorum-part2.html
Please reply me if anything is not