请教各位老师一个查询缓慢的问题

[SKYlwc912]

我这因为要做一个统计，所以写了这么一个sql语句：
select * from (select decode(a.Status,0,'受理',1,'承办环节',2,'审核环节',3,'批准环节',4,'办结环节',100,'办结',a.Status) as Status_NAME,
a.Status,a.nid,to_char(a.ISSUE_DATE,'YYYY-MM-DD') as ISSUE_DATE,a.FILE_ID,a.APP_UINT,a.APP_INFO,a.TRANSACT,
nvl(c.UserNAme,a.TRANSACT) as CURRENT_TRANSACT,a.SHENPISHIXIAN,b.name,nvl(d.bjjg,1000) as bjjg,a.EXAM_TYPE,a.ADDRESS,a.JLR,a.IS_BIGPRJ,
a.RETURN_RES,e.NAME as BJJG_NAME from EXAM_ACCEPT a,EXAM_DIC_FLOW b,
(select e.*,g.UserNAme from EXAM_LOG e,Sys_Users g where e.RECEIVER = g.usercode and NID=(select max(NID) from EXAM_LOG f where e.FILE_NUMBER=f.FILE_NUMBER)) c,
EXAM_END d,(select * from SYS_DIC where DIC_TYPE=3) e
where a.Exam_Type=b.opration_code and a.nid=c.FILE_NUMBER(+) and a.NID=d.file_id(+) and d.bjjg=e.NID(+) and a.TAG_DEL<>1 ) t where 1=1
order by t.Issue_Date desc
但是它很耗费时间，这几个表里最多记录的也就2万多条记录，但这语句一执行就要差不多一小时才出结果，造成慢的语句是第5行搜索max(NID)的那部分特费时间。不知道有什么办法优化一下不？
我试过把*换成具体列名也还是一样慢。

[oracledbalgtu]

 创建如下index,并修改代码如下:
 CREATE INDEX i_EXAM_LOG_aaa ON EXAM_LOG(FILE_NUMBER,MNID DESC);

SELECT *
  FROM (SELECT DECODE(A.STATUS,
                      0,
                      '受理',
                      1,
                      '承办环节',
                      2,
                      '审核环节',
                      3,
                      '批准环节',
                      4,
                      '办结环节',
                      100,
                      '办结',
                      A.STATUS) AS STATUS_NAME,
               A.STATUS,
               A.NID,
               TO_CHAR(A.ISSUE_DATE, 'YYYY-MM-DD') AS ISSUE_DATE,
               A.FILE_ID,
               A.APP_UINT,
               A.APP_INFO,
               A.TRANSACT,
               NVL(C.USERNAME, A.TRANSACT) AS CURRENT_TRANSACT,
               A.SHENPISHIXIAN,
               B.NAME,
               NVL(D.BJJG, 1000) AS BJJG,
               A.EXAM_TYPE,
               A.ADDRESS,
               A.JLR,
               A.IS_BIGPRJ,
               A.RETURN_RES,
               E.NAME AS BJJG_NAME
          FROM EXAM_ACCEPT A,
               EXAM_DIC_FLOW B,
               (SELECT E.*, G.USERNAME
                  FROM EXAM_LOG E,
                       SYS_USERS G,
                       (SELECT MAX(NID) MNID, FILE_NUMBER
                          FROM EXAM_LOG
                         GROUP BY FILE_NUMBER) X
                 WHERE E.RECEIVER = G.USERCODE
                   AND E.NID = X.MNID
                   AND E.FILE_NUMBER = X.FILE_NUMBER) C,
               EXAM_END D,
               (SELECT * FROM SYS_DIC WHERE DIC_TYPE = 3) E
         WHERE A.EXAM_TYPE = B.OPRATION_CODE
           AND A.NID = C.FILE_NUMBER(+)
           AND A.NID = D.FILE_ID(+)
           AND D.BJJG = E.NID(+)
           AND A.TAG_DEL <> 1) T
 WHERE 1 = 1
 ORDER BY T.ISSUE_DATE DESC;
 

[Quote=引用楼主 SKYlwc912 的帖子:]
我这因为要做一个统计，所以写了这么一个sql语句：
select * from (select decode(a.Status,0,'受理',1,'承办环节',2,'审核环节',3,'批准环节',4,'办结环节',100,'办结',a.Status) as Status_NAME,
a.Status,a.nid,to_char(a.ISSUE_DATE,'YYYY-MM-DD') as ISSUE_DATE,a.FILE_ID,a.APP_UINT,a.APP_INFO,a.TRANSACT,
nvl(c.UserNAme,a.TRANSACT) as CURRENT_TRANSACT,a.SHENPISHIXIAN,b.name,nvl(d.bjjg,1000) as bjjg,a.EX…
[/Quote]

[SKYlwc912]

CREATE INDEX i_EXAM_LOG_aaa ON EXAM_LOG(FILE_NUMBER,MNID DESC);
中的MNID 要先声明一下吗，创建时报错“ORA-00904: "MNID": 无效的标识符”

[SKYlwc912]

多谢oracledbalgtu，我把语句改了就可以直接用了，而且速度很快，非常感谢！