diff --git a/business/views.py b/business/views.py
index 2f8a00b..974e773 100644
--- a/business/views.py
+++ b/business/views.py
@@ -4246,8 +4246,23 @@ def _extract_all_person_info(info_val):
                     originals.append(orig)
             if names or id_numbers:
                 return names, id_numbers, originals
-    # 回退到 parse_person_info 处理普通字符串
+    # 回退到处理普通字符串
     if isinstance(info_val, str):
+        # 先尝试匹配多个 "name(id)" 格式（如 "张三(123456),李四(789012)"）
+        matches = re.findall(r'([^,，（(]+?)\s*[（(]([^）)]*)[）)]', info_val)
+        if matches:
+            for name, id_num in matches:
+                name = name.strip()
+                id_num = id_num.strip()
+                if name:
+                    names.append(name)
+                if id_num:
+                    id_numbers.append(id_num)
+                orig = f"{name}（{id_num}）" if name and id_num else (name or id_num or '')
+                if orig:
+                    originals.append(orig)
+            return names, id_numbers, originals
+        # 没有匹配到多人格式，回退到单人解析
         parsed = parse_person_info(info_val)
         if parsed['name']:
             names.append(parsed['name'])
@@ -4656,12 +4671,22 @@ class ConflictSearch(APIView):
         )
         # 归一化冲突记录为前端期望格式 [{ index, name, idNumber }, ...]（预立案用 client_username/party_username，立项用 client_info/party_info，投标用 BiddingUnit）
         def _normalize_person_list(val):
+            """
+            归一化人员信息，始终返回 list[dict] 或 None。
+            支持：
+              - JSON 数组字符串：'[{"name":"张三","idNumber":"111"}]'
+              - JSON 对象字符串：'{"name":"张三","idNumber":"111"}'
+              - list / dict（已解析的 JSON）
+              - 纯字符串："张三（111）"、"张三(111),李四(222)"
+            """
             if not val:
                 return []
             try:
                 lst = json.loads(val) if isinstance(val, str) else val
+                if isinstance(lst, dict):
+                    lst = [lst]
                 if not isinstance(lst, list):
-                    return None
+                    raise ValueError("not a list")
                 out = []
                 for i, item in enumerate(lst):
                     if not isinstance(item, dict):
@@ -4679,9 +4704,37 @@ class ConflictSearch(APIView):
                     })
                 return out
             except (json.JSONDecodeError, TypeError, ValueError):
-                return None
+                pass
+
+            # 回退：解析纯字符串格式（如 "张三（123456）" 或 "张三(123),李四(456)"）
+            if isinstance(val, str):
+                # 尝试匹配多个 "name(id)" 格式
+                matches = re.findall(r'([^,，（(]+?)\s*[（(]([^）)]*)[）)]', val)
+                if matches:
+                    out = []
+                    for i, (name, id_num) in enumerate(matches):
+                        out.append({
+                            'index': i + 1,
+                            'name': name.strip(),
+                            'idNumber': id_num.strip()
+                        })
+                    return out
+                # 没有括号格式，尝试用 parse_person_info 解析
+                parsed = parse_person_info(val)
+                if parsed['name'] or parsed['id_number']:
+                    return [{
+                        'index': 1,
+                        'name': parsed['name'],
+                        'idNumber': parsed['id_number']
+                    }]
+            return None
 
         def _normalize_conflict_records(records):
+            """
+            归一化冲突记录，将人员信息字段统一转为 JSON 字符串格式：
+            '[{"index":1,"name":"张三","idNumber":"123456"}]'
+            确保前端可以使用 isValidFormat() + JSON.parse() 正确展示。
+            """
             processed = []
             for record in records:
                 new_record = dict(record)
@@ -4690,7 +4743,8 @@ class ConflictSearch(APIView):
                         continue
                     normalized = _normalize_person_list(new_record[key])
                     if normalized is not None:
-                        new_record[key] = normalized
+                        # 转为 JSON 字符串，兼容前端 JSON.parse() 展示
+                        new_record[key] = json.dumps(normalized, ensure_ascii=False)
                 processed.append(new_record)
             return processed