一个语句慢，如何更优化。。。。。急
SQL：
Select   count(siim.siim_imp_bl_id)   as   blcount,
              sum(decode(siim.siim_transfer_flag,   'Y ',   1,   0))   as   transblcount,
              sum(spgi.spgi_quantity)   as   cargoweight,
              sum(decode(siim.siim_transfer_flag,   'Y ',   spgi.spgi_quantity,   0))   as   transcargoweight
    from   (select   b.spgi_imp_bl_id,   sum(b.SPGI_gross_weight)   /   1000   spgi_quantity
                    from   scargoinfo   b
                  where   b.spgi_record_type   =   '1 '
                  group   by   b.spgi_imp_bl_id)   spgi,
              cshipcanonical   cshc,
              shsailingschedule   shss,
                  simportmanifest   siim
  where   siim.siim_voyage_id   =   shss.shss_voyage_id(+)  
      and   shss.shss_vessel_code   =   cshc.cshc_id(+)  
      and   siim.siim_imp_bl_id   =   spgi.spgi_imp_bl_id(+)  
      and   siim.siim_valid_flag   =   'Y '  
      and   siim.siim_voyage_id   =   '2683 '  
      and   siim.siim_carrier_id   =   'ESSA00100415 '  
      and   shss.shss_eta_time   > =   to_date( '2006-01-01 ',   'yyyy-mm-dd   hh24:mi:ss ')
      and   shss.shss_eta_time   <=   to_date( '2007-12-12 ',   'yyyy-mm-dd   hh24:mi:ss ')  
      and   shss.shss_berthing_time   > =   to_date( '2006-01-01 ',   'yyyy-mm-dd   hh24:mi:ss ')  
      and   shss.shss_berthing_time   <=   to_date( '2007-12-12 ',   'yyyy-mm-dd   hh24:mi:ss ')  
      and   SIIM.SIIM_ORG_ID   =   '139 '
该语句结果为1条记录，用时1.7S，如果把
(select   b.spgi_imp_bl_id,   sum(b.SPGI_gross_weight)   /   1000   spgi_quantity
                    from   scargoinfo   b
                  where   b.spgi_record_type   =   '1 '
                  group   by   b.spgi_imp_bl_id)   spgi   去掉，用时为0.012s，我对scargoinfo的spgi_imp_bl_id，spgi_record_type，SPGI_gross_weight建立了联合索引，还是要0.84s，请问各位有更好的解决方案吗。



------解决方案--------------------
单独执行
select b.spgi_imp_bl_id, sum(b.SPGI_gross_weight) / 1000 spgi_quantity
from scargoinfo b
where b.spgi_record_type = '1 '
group by b.spgi_imp_bl_id
要用多长时间，返回多少记录。