diff --git a/src/main/java/com/kamco/cd/kamcoback/inference/dto/InferenceResultDto.java b/src/main/java/com/kamco/cd/kamcoback/inference/dto/InferenceResultDto.java
index fecbb7e1..6ab2556b 100644
--- a/src/main/java/com/kamco/cd/kamcoback/inference/dto/InferenceResultDto.java
+++ b/src/main/java/com/kamco/cd/kamcoback/inference/dto/InferenceResultDto.java
@@ -23,7 +23,9 @@ import org.springframework.data.domain.Pageable;
 
 public class InferenceResultDto {
 
-  /** 목록조회 dto */
+  /**
+   * 목록조회 dto
+   */
   @Getter
   @Setter
   @AllArgsConstructor
@@ -35,11 +37,15 @@ public class InferenceResultDto {
     private String status;
     private String mapSheetCnt;
     private Long detectingCnt;
-    @JsonFormatDttm private ZonedDateTime startTime;
-    @JsonFormatDttm private ZonedDateTime endTime;
-    @JsonFormatDttm private ZonedDateTime elapsedTime;
+    @JsonFormatDttm
+    private ZonedDateTime startTime;
+    @JsonFormatDttm
+    private ZonedDateTime endTime;
+    @JsonFormatDttm
+    private ZonedDateTime elapsedTime;
     private Boolean applyYn;
-    @JsonFormatDttm private ZonedDateTime applyDttm;
+    @JsonFormatDttm
+    private ZonedDateTime applyDttm;
 
     @JsonProperty("statusName")
     public String statusName() {
@@ -47,7 +53,9 @@ public class InferenceResultDto {
     }
   }
 
-  /** 목록조회 검색 조건 dto */
+  /**
+   * 목록조회 검색 조건 dto
+   */
   @Getter
   @Setter
   @NoArgsConstructor
@@ -69,7 +77,9 @@ public class InferenceResultDto {
     }
   }
 
-  /** 탐지 데이터 옵션 dto */
+  /**
+   * 탐지 데이터 옵션 dto
+   */
   @Getter
   @AllArgsConstructor
   public enum MapSheetScope implements EnumType {
@@ -90,7 +100,9 @@ public class InferenceResultDto {
     }
   }
 
-  /** 분석대상 도엽 enum */
+  /**
+   * 분석대상 도엽 enum
+   */
   @Getter
   @AllArgsConstructor
   public enum DetectOption implements EnumType {
@@ -138,7 +150,9 @@ public class InferenceResultDto {
     }
   }
 
-  /** 변화탐지 실행 정보 저장 요청 정보 */
+  /**
+   * 변화탐지 실행 정보 저장 요청 정보
+   */
   @Getter
   @Setter
   @NoArgsConstructor
@@ -177,8 +191,8 @@ public class InferenceResultDto {
     @Schema(description = "탐지 데이터 옵션 - 추론제외(EXCL), 이전 년도 도엽 사용(PREV)", example = "EXCL")
     @NotBlank
     @EnumValid(
-        enumClass = DetectOption.class,
-        message = "탐지 데이터 옵션은 '추론제외', '이전 년도 도엽 사용' 만 사용 가능합니다.")
+      enumClass = DetectOption.class,
+      message = "탐지 데이터 옵션은 '추론제외', '이전 년도 도엽 사용' 만 사용 가능합니다.")
     private String detectOption;
 
     @Schema(description = "5k 도협 번호 목록", example = "[33605,33606, 33610, 34802, 35603, 35611]")
@@ -205,41 +219,49 @@ public class InferenceResultDto {
     private Integer targetYyyy;
     private String detectOption;
     private String mapSheetScope;
-    @JsonFormatDttm private ZonedDateTime inferStartDttm;
-    @JsonFormatDttm private ZonedDateTime inferEndDttm;
+    @JsonFormatDttm
+    private ZonedDateTime inferStartDttm;
+    @JsonFormatDttm
+    private ZonedDateTime inferEndDttm;
     private Long detectingCnt = 0L;
     private Long detectingEndCnt = 0L;
 
-    @JsonFormatDttm private ZonedDateTime m1ModelStartDttm;
-    @JsonFormatDttm private ZonedDateTime m1ModelEndDttm;
-    @JsonFormatDttm private ZonedDateTime m2ModelStartDttm;
-    @JsonFormatDttm private ZonedDateTime m2ModelEndDttm;
-    @JsonFormatDttm private ZonedDateTime m3ModelStartDttm;
-    @JsonFormatDttm private ZonedDateTime m3ModelEndDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m1ModelStartDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m1ModelEndDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m2ModelStartDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m2ModelEndDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m3ModelStartDttm;
+    @JsonFormatDttm
+    private ZonedDateTime m3ModelEndDttm;
 
     private String model1Ver;
     private String model2Ver;
     private String model3Ver;
 
     public InferenceStatusDetailDto(
-        String title,
-        Integer compareYyyy,
-        Integer targetYyyy,
-        String detectOption,
-        String mapSheetScope,
-        ZonedDateTime inferStartDttm,
-        ZonedDateTime inferEndDttm,
-        Long detectingCnt,
-        Long detectingEndCnt,
-        ZonedDateTime m1ModelStartDttm,
-        ZonedDateTime m1ModelEndDttm,
-        ZonedDateTime m2ModelStartDttm,
-        ZonedDateTime m2ModelEndDttm,
-        ZonedDateTime m3ModelStartDttm,
-        ZonedDateTime m3ModelEndDttm,
-        String model1Ver,
-        String model2Ver,
-        String model3Ver) {
+      String title,
+      Integer compareYyyy,
+      Integer targetYyyy,
+      String detectOption,
+      String mapSheetScope,
+      ZonedDateTime inferStartDttm,
+      ZonedDateTime inferEndDttm,
+      Long detectingCnt,
+      Long detectingEndCnt,
+      ZonedDateTime m1ModelStartDttm,
+      ZonedDateTime m1ModelEndDttm,
+      ZonedDateTime m2ModelStartDttm,
+      ZonedDateTime m2ModelEndDttm,
+      ZonedDateTime m3ModelStartDttm,
+      ZonedDateTime m3ModelEndDttm,
+      String model1Ver,
+      String model2Ver,
+      String model3Ver) {
       this.title = title;
       this.compareYyyy = compareYyyy;
       this.targetYyyy = targetYyyy;
@@ -342,9 +364,12 @@ public class InferenceResultDto {
   public static class InferenceServerStatusDto {
 
     private String serverName;
-    @JsonIgnore private float cpu_user;
-    @JsonIgnore private float cpu_system;
-    @JsonIgnore private float memused;
+    @JsonIgnore
+    private float cpu_user;
+    @JsonIgnore
+    private float cpu_system;
+    @JsonIgnore
+    private float memused;
     private Long kbmemused;
     private float gpuUtil;
 
@@ -458,5 +483,9 @@ public class InferenceResultDto {
     private ZonedDateTime modelEndDttm;
     private Long updateUid;
     private String runningModelType;
+    private Integer pendingJobs;
+    private Integer runningJobs;
+    private Integer completedJobs;
+    private Integer failedJobs;
   }
 }
diff --git a/src/main/java/com/kamco/cd/kamcoback/inference/service/InferenceResultService.java b/src/main/java/com/kamco/cd/kamcoback/inference/service/InferenceResultService.java
index 617b8bbc..ff31f90c 100644
--- a/src/main/java/com/kamco/cd/kamcoback/inference/service/InferenceResultService.java
+++ b/src/main/java/com/kamco/cd/kamcoback/inference/service/InferenceResultService.java
@@ -106,7 +106,7 @@ public class InferenceResultService {
 
     // 변화탐지 실행 가능 비교년도 조회
     List<MngListCompareDto> compareList =
-        mapSheetMngCoreService.getByHstMapSheetCompareList(req.getCompareYyyy(), mapTargetIds);
+      mapSheetMngCoreService.getByHstMapSheetCompareList(req.getCompareYyyy(), mapTargetIds);
 
     if (compareList.isEmpty()) {
       throw new CustomApiException("NOT_FOUND_COMPARE_YEAR", HttpStatus.NOT_FOUND);
@@ -140,6 +140,7 @@ public class InferenceResultService {
       throw new CustomApiException("NOT_FOUND_COMPARE_YEAR", HttpStatus.NOT_FOUND);
     }
 
+    // 사용할 영상파일 년도 기록 및 추론에 포함되는지 설정
     for (MngListDto target : targetList) {
       for (Map<String, Object> map : totalNumList) {
         if (target.getMapSheetNum().equals(map.get("mapSheetNum").toString())) {
@@ -154,20 +155,20 @@ public class InferenceResultService {
 
     // 추론에 필요한 geojson 파일 생성
     List<String> mapSheetNumList =
-        targetList.stream()
-            .filter(t -> Boolean.TRUE.equals(t.getIsSuccess()))
-            .map(MngListDto::getMapSheetNum)
-            .toList();
+      targetList.stream()
+        .filter(t -> Boolean.TRUE.equals(t.getIsSuccess()))
+        .map(MngListDto::getMapSheetNum)
+        .toList();
 
     // 비교년도 geojson 파일 생성하여 경로 받기
     String modelComparePath =
-        getSceneInference(
-            String.valueOf(req.getCompareYyyy()), mapSheetNumList, req.getMapSheetScope());
+      getSceneInference(
+        String.valueOf(req.getCompareYyyy()), mapSheetNumList, req.getMapSheetScope());
 
     // 기준년도 geojson 파일 생성하여 경로 받기
     String modelTargetPath =
-        getSceneInference(
-            String.valueOf(req.getTargetYyyy()), mapSheetNumList, req.getMapSheetScope());
+      getSceneInference(
+        String.valueOf(req.getTargetYyyy()), mapSheetNumList, req.getMapSheetScope());
 
     // ai 서버에 전달할 파라미터 생성
     pred_requests_areas predRequestsAreas = new pred_requests_areas();
@@ -199,19 +200,19 @@ public class InferenceResultService {
 
   // 비교년도 탐지 제이터 옵션 별로 조회하여 req에 적용
   private List<MapSheetNumDto> createdMngDto(
-      InferenceResultDto.RegReq req, List<MngListDto> targetList) {
+    InferenceResultDto.RegReq req, List<MngListDto> targetList) {
     List<String> mapTargetIds = new ArrayList<>();
 
     targetList.forEach(
-        hstMapSheet -> {
-          // 비교년도는 target 년도 기준으로 가져옴 파라미터 만들기
-          mapTargetIds.add(hstMapSheet.getMapSheetNum());
-        });
+      hstMapSheet -> {
+        // 비교년도는 target 년도 기준으로 가져옴 파라미터 만들기
+        mapTargetIds.add(hstMapSheet.getMapSheetNum());
+      });
 
     // 비교년도 조회
     List<String> mapCompareIds = new ArrayList<>();
     List<MngListCompareDto> compareList =
-        mapSheetMngCoreService.getByHstMapSheetCompareList(req.getCompareYyyy(), mapTargetIds);
+      mapSheetMngCoreService.getByHstMapSheetCompareList(req.getCompareYyyy(), mapTargetIds);
 
     for (MngListCompareDto dto : compareList) {
       // 추론 제외일때 이전년도 파일이 없으면 제외
@@ -226,35 +227,35 @@ public class InferenceResultService {
     }
 
     Set<String> compareSet =
-        mapCompareIds.stream()
-            .filter(Objects::nonNull)
-            .map(String::trim) // 공백/개행 방지
-            .collect(Collectors.toSet());
+      mapCompareIds.stream()
+        .filter(Objects::nonNull)
+        .map(String::trim) // 공백/개행 방지
+        .collect(Collectors.toSet());
 
     // target 기준 compare 비교하여 서로 있는것만 저장
     List<String> commonIds =
-        mapTargetIds.stream()
-            .filter(Objects::nonNull)
-            .map(String::trim)
-            .filter(compareSet::contains)
-            .toList();
+      mapTargetIds.stream()
+        .filter(Objects::nonNull)
+        .map(String::trim)
+        .filter(compareSet::contains)
+        .toList();
 
     Set<String> commonIdSet =
-        commonIds.stream().filter(Objects::nonNull).map(String::trim).collect(Collectors.toSet());
+      commonIds.stream().filter(Objects::nonNull).map(String::trim).collect(Collectors.toSet());
 
     // 저장하기위해 파라미터 다시 구성
     List<MapSheetNumDto> mapSheetNum =
-        targetList.stream()
-            .filter(dto -> dto.getMapSheetNum() != null)
-            .filter(dto -> commonIdSet.contains(dto.getMapSheetNum().trim()))
-            .map(
-                dto -> {
-                  MapSheetNumDto mapSheetNumDto = new MapSheetNumDto();
-                  mapSheetNumDto.setMapSheetNum(dto.getMapSheetNum());
-                  mapSheetNumDto.setMapSheetName(dto.getMapSheetName());
-                  return mapSheetNumDto;
-                })
-            .toList();
+      targetList.stream()
+        .filter(dto -> dto.getMapSheetNum() != null)
+        .filter(dto -> commonIdSet.contains(dto.getMapSheetNum().trim()))
+        .map(
+          dto -> {
+            MapSheetNumDto mapSheetNumDto = new MapSheetNumDto();
+            mapSheetNumDto.setMapSheetNum(dto.getMapSheetNum());
+            mapSheetNumDto.setMapSheetName(dto.getMapSheetName());
+            return mapSheetNumDto;
+          })
+        .toList();
 
     return mapSheetNum;
   }
@@ -271,14 +272,14 @@ public class InferenceResultService {
       throw new CustomApiException("BAD_REQUEST", HttpStatus.BAD_REQUEST);
     }
 
-    // 1) 요청 로그 (debug 권장)
+    // 1) 요청 로그
     try {
       log.debug("Inference request dto={}", objectMapper.writeValueAsString(dto));
     } catch (JsonProcessingException e) {
       log.warn("Failed to serialize inference dto", e);
     }
 
-    // 2) local 환경 임시 처리 (NPE 방어)
+    // 2) local 환경 임시 처리
     if ("local".equals(profile)) {
       if (dto.getPred_requests_areas() == null) {
         throw new IllegalStateException("pred_requests_areas is null");
@@ -293,7 +294,7 @@ public class InferenceResultService {
     headers.setAccept(List.of(MediaType.APPLICATION_JSON));
 
     ExternalCallResult<String> result =
-        externalHttpClient.call(inferenceUrl, HttpMethod.POST, dto, headers, String.class);
+      externalHttpClient.call(inferenceUrl, HttpMethod.POST, dto, headers, String.class);
 
     if (result.statusCode() < 200 || result.statusCode() >= 300) {
       log.error("Inference API failed. status={}, body={}", result.statusCode(), result.body());
@@ -303,7 +304,8 @@ public class InferenceResultService {
     // 4) 응답 파싱
     try {
       List<Map<String, Object>> list =
-          objectMapper.readValue(result.body(), new TypeReference<>() {});
+        objectMapper.readValue(result.body(), new TypeReference<>() {
+        });
 
       if (list.isEmpty()) {
         throw new IllegalStateException("Inference response is empty");
@@ -372,8 +374,8 @@ public class InferenceResultService {
   /**
    * geojson 파일 생성
    *
-   * @param yyyy 영상관리 파일별 년도
-   * @param mapSheetNums 5k 도엽 번호 리스트
+   * @param yyyy          영상관리 파일별 년도
+   * @param mapSheetNums  5k 도엽 번호 리스트
    * @param mapSheetScope EXCL : 추론제외, PREV 이전 년도 도엽 사용
    * @return
    */
@@ -408,7 +410,7 @@ public class InferenceResultService {
    * @return
    */
   public Page<InferenceDetailDto.Geom> getInferenceResultGeomList(
-      Long id, InferenceDetailDto.SearchGeoReq searchGeoReq) {
+    Long id, InferenceDetailDto.SearchGeoReq searchGeoReq) {
     return inferenceResultCoreService.getInferenceResultGeomList(id, searchGeoReq);
   }
 
@@ -419,7 +421,7 @@ public class InferenceResultService {
    * @return
    */
   public Page<InferenceDetailDto.DetailListEntity> listInferenceResultWithGeom(
-      @NotNull Long id, InferenceDetailDto.SearchGeoReq searchReq) {
+    @NotNull Long id, InferenceDetailDto.SearchGeoReq searchReq) {
 
     return inferenceResultCoreService.listInferenceResultWithGeom(id, searchReq);
   }
@@ -464,7 +466,7 @@ public class InferenceResultService {
   public InferenceStatusDetailDto getInferenceStatus(UUID uuid) {
 
     List<InferenceServerStatusDto> servers =
-        inferenceResultCoreService.getInferenceServerStatusList();
+      inferenceResultCoreService.getInferenceServerStatusList();
 
     String serverNames = "";
     for (InferenceServerStatusDto server : servers) {
diff --git a/src/main/java/com/kamco/cd/kamcoback/postgres/core/InferenceResultCoreService.java b/src/main/java/com/kamco/cd/kamcoback/postgres/core/InferenceResultCoreService.java
index a7347cdd..5967dcf0 100644
--- a/src/main/java/com/kamco/cd/kamcoback/postgres/core/InferenceResultCoreService.java
+++ b/src/main/java/com/kamco/cd/kamcoback/postgres/core/InferenceResultCoreService.java
@@ -29,6 +29,7 @@ import java.time.ZonedDateTime;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.UUID;
+import java.util.function.Consumer;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.log4j.Log4j2;
 import org.springframework.dao.DataAccessException;
@@ -101,7 +102,7 @@ public class InferenceResultCoreService {
     mapSheetLearnEntity.setMapSheetCnt(mapSheetName);
     mapSheetLearnEntity.setDetectingCnt((long) detectingCnt);
     mapSheetLearnEntity.setStage(
-        mapSheetLearnRepository.getLearnStage(req.getCompareYyyy(), req.getTargetYyyy()));
+      mapSheetLearnRepository.getLearnStage(req.getCompareYyyy(), req.getTargetYyyy()));
 
     // learn 테이블 저장
     MapSheetLearnEntity savedLearn = mapSheetLearnRepository.save(mapSheetLearnEntity);
@@ -163,9 +164,9 @@ public class InferenceResultCoreService {
    */
   public InferenceDetailDto.AnalResSummary getInferenceResultSummary(Long id) {
     InferenceDetailDto.AnalResSummary summary =
-        mapSheetAnalDataRepository
-            .getInferenceResultSummary(id)
-            .orElseThrow(() -> new EntityNotFoundException("요약정보를 찾을 수 없습니다. " + id));
+      mapSheetAnalDataRepository
+        .getInferenceResultSummary(id)
+        .orElseThrow(() -> new EntityNotFoundException("요약정보를 찾을 수 없습니다. " + id));
     return summary;
   }
 
@@ -186,7 +187,7 @@ public class InferenceResultCoreService {
    * @return
    */
   public Page<InferenceDetailDto.Geom> getInferenceResultGeomList(
-      Long id, InferenceDetailDto.SearchGeoReq searchGeoReq) {
+    Long id, InferenceDetailDto.SearchGeoReq searchGeoReq) {
     return mapSheetAnalDataRepository.getInferenceGeomList(id, searchGeoReq);
   }
 
@@ -198,16 +199,16 @@ public class InferenceResultCoreService {
    */
   @Transactional(readOnly = true)
   public Page<InferenceDetailDto.DetailListEntity> listInferenceResultWithGeom(
-      @NotNull Long analyId, InferenceDetailDto.SearchGeoReq searchReq) {
+    @NotNull Long analyId, InferenceDetailDto.SearchGeoReq searchReq) {
     // 분석 ID 에 해당하는 dataids를 가져온다.
     List<Long> dataIds =
-        mapSheetAnalDataRepository.listAnalyGeom(analyId).stream()
-            .mapToLong(MapSheetAnalDataInferenceEntity::getId)
-            .boxed()
-            .toList();
+      mapSheetAnalDataRepository.listAnalyGeom(analyId).stream()
+        .mapToLong(MapSheetAnalDataInferenceEntity::getId)
+        .boxed()
+        .toList();
     // 해당데이터의 폴리곤데이터를 가져온다
     Page<MapSheetAnalDataInferenceGeomEntity> mapSheetAnalDataGeomEntities =
-        mapSheetAnalDataRepository.listInferenceResultWithGeom(dataIds, searchReq);
+      mapSheetAnalDataRepository.listInferenceResultWithGeom(dataIds, searchReq);
     return mapSheetAnalDataGeomEntities.map(MapSheetAnalDataInferenceGeomEntity::toEntity);
   }
 
@@ -224,108 +225,112 @@ public class InferenceResultCoreService {
   @Transactional(readOnly = true)
   public List<MapSheet> listGetScenes5k(Long analyId) {
     List<String> sceneCodes =
-        mapSheetAnalDataRepository.listAnalyGeom(analyId).stream()
-            .mapToLong(MapSheetAnalDataInferenceEntity::getMapSheetNum)
-            .mapToObj(String::valueOf)
-            .toList();
+      mapSheetAnalDataRepository.listAnalyGeom(analyId).stream()
+        .mapToLong(MapSheetAnalDataInferenceEntity::getMapSheetNum)
+        .mapToObj(String::valueOf)
+        .toList();
 
     return mapInkx5kRepository.listGetScenes5k(sceneCodes).stream()
-        .map(MapInkx5kEntity::toEntity)
-        .toList();
+      .map(MapInkx5kEntity::toEntity)
+      .toList();
   }
 
+  /**
+   * learn 테이블 update
+   *
+   * @param request 추론 실행 정보
+   */
   public void update(SaveInferenceAiDto request) {
     MapSheetLearnEntity entity =
-        mapSheetLearnRepository
-            .getInferenceResultByUuid(request.getUuid())
-            .orElseThrow(() -> new EntityNotFoundException());
+      mapSheetLearnRepository.getInferenceResultByUuid(request.getUuid())
+        .orElseThrow(EntityNotFoundException::new);
 
+    // M1/M2/M3 영역 업데이트
     if (request.getType() != null) {
-      switch (request.getType()) {
-        case "M1" -> {
-          if (request.getBatchId() != null) {
-            entity.setM1ModelBatchId(request.getBatchId());
-          }
-
-          if (request.getModelStartDttm() != null) {
-            entity.setM1ModelStartDttm(request.getModelStartDttm());
-          }
-
-          if (request.getModelEndDttm() != null) {
-            entity.setM1ModelEndDttm(request.getModelEndDttm());
-          }
-        }
-        case "M2" -> {
-          if (request.getBatchId() != null) {
-            entity.setM2ModelBatchId(request.getBatchId());
-          }
-
-          if (request.getModelStartDttm() != null) {
-            entity.setM2ModelStartDttm(request.getModelStartDttm());
-          }
-
-          if (request.getModelEndDttm() != null) {
-            entity.setM2ModelEndDttm(request.getModelEndDttm());
-          }
-        }
-        case "M3" -> {
-          if (request.getBatchId() != null) {
-            entity.setM3ModelBatchId(request.getBatchId());
-          }
-
-          if (request.getModelStartDttm() != null) {
-            entity.setM3ModelStartDttm(request.getModelStartDttm());
-          }
-
-          if (request.getModelEndDttm() != null) {
-            entity.setM3ModelEndDttm(request.getModelEndDttm());
-          }
-        }
-      }
+      applyModelUpdate(entity, request);
     }
 
-    if (request.getRunningModelType() != null) {
-      entity.setRunningModelType(request.getRunningModelType());
-    }
-
-    if (request.getInferStartDttm() != null) {
-      entity.setInferStartDttm(request.getInferStartDttm());
-    }
-
-    if (request.getInferEndDttm() != null) {
-      entity.setInferEndDttm(request.getInferEndDttm());
-    }
-
-    if (request.getModelComparePath() != null) {
-      entity.setModelComparePath(request.getModelComparePath());
-    }
-
-    if (request.getModelTargetPath() != null) {
-      entity.setModelTargetPath(request.getModelTargetPath());
-    }
-
-    if (request.getDetectEndCnt() != null) {
-      entity.setDetectEndCnt(request.getDetectEndCnt());
-    }
-
-    if (request.getStatus() != null) {
-      entity.setStatus(request.getStatus());
-    }
-
-    if (request.getUpdateUid() != null) {
-      entity.setUpdatedUid(request.getUpdateUid());
-    }
+    // 공통 영역 업데이트
+    applyIfNotNull(request.getRunningModelType(), entity::setRunningModelType);
+    applyIfNotNull(request.getInferStartDttm(), entity::setInferStartDttm);
+    applyIfNotNull(request.getInferEndDttm(), entity::setInferEndDttm);
+    applyIfNotNull(request.getModelComparePath(), entity::setModelComparePath);
+    applyIfNotNull(request.getModelTargetPath(), entity::setModelTargetPath);
+    applyIfNotNull(request.getDetectEndCnt(), entity::setDetectEndCnt);
+    applyIfNotNull(request.getStatus(), entity::setStatus);
+    applyIfNotNull(request.getUpdateUid(), entity::setUpdatedUid);
 
     entity.setUpdatedDttm(ZonedDateTime.now());
   }
 
+  private void applyModelUpdate(MapSheetLearnEntity entity, SaveInferenceAiDto request) {
+    switch (request.getType()) {
+      case "M1" -> applyModelFields(
+        request,
+        entity::setM1ModelBatchId,
+        entity::setM1ModelStartDttm,
+        entity::setM1ModelEndDttm,
+        entity::setM1PendingJobs,
+        entity::setM1RunningJobs,
+        entity::setM1CompletedJobs,
+        entity::setM1FailedJobs
+      );
+      case "M2" -> applyModelFields(
+        request,
+        entity::setM2ModelBatchId,
+        entity::setM2ModelStartDttm,
+        entity::setM2ModelEndDttm,
+        entity::setM2PendingJobs,
+        entity::setM2RunningJobs,
+        entity::setM2CompletedJobs,
+        entity::setM2FailedJobs
+      );
+      case "M3" -> applyModelFields(
+        request,
+        entity::setM3ModelBatchId,
+        entity::setM3ModelStartDttm,
+        entity::setM3ModelEndDttm,
+        entity::setM3PendingJobs,
+        entity::setM3RunningJobs,
+        entity::setM3CompletedJobs,
+        entity::setM3FailedJobs
+      );
+      default -> throw new IllegalArgumentException("Unknown type: " + request.getType());
+    }
+  }
+
+  private void applyModelFields(
+    SaveInferenceAiDto request,
+    Consumer<Long> setBatchId,
+    Consumer<ZonedDateTime> setStart,
+    Consumer<ZonedDateTime> setEnd,
+    Consumer<Integer> setPending,
+    Consumer<Integer> setRunning,
+    Consumer<Integer> setCompleted,
+    Consumer<Integer> setFailed
+  ) {
+    applyIfNotNull(request.getBatchId(), setBatchId);
+    applyIfNotNull(request.getModelStartDttm(), setStart);
+    applyIfNotNull(request.getModelEndDttm(), setEnd);
+    applyIfNotNull(request.getPendingJobs(), setPending);
+    applyIfNotNull(request.getRunningJobs(), setRunning);
+    applyIfNotNull(request.getCompletedJobs(), setCompleted);
+    applyIfNotNull(request.getFailedJobs(), setFailed);
+  }
+
+  private <T> void applyIfNotNull(T value, Consumer<T> setter) {
+    if (value != null) {
+      setter.accept(value);
+    }
+  }
+
   public List<InferenceServerStatusDto> getInferenceServerStatusList() {
     return mapSheetLearnRepository.getInferenceServerStatusList();
   }
 
   public InferenceBatchSheet getInferenceResultByStatus(String status) {
     MapSheetLearnEntity entity =
-        mapSheetLearnRepository.getInferenceResultByStatus(status).orElse(null);
+      mapSheetLearnRepository.getInferenceResultByStatus(status).orElse(null);
 
     if (entity == null) {
       return null;
@@ -345,7 +350,14 @@ public class InferenceResultCoreService {
     return inferenceBatchSheet;
   }
 
-  public InferenceProgressDto getInferenceAiResultById(Long id, String type, UUID modelUuid) {
+  /**
+   * 추론 실행 api 파라미터 조회
+   *
+   * @param id
+   * @param modelUuid
+   * @return
+   */
+  public InferenceProgressDto getInferenceAiResultById(Long id, UUID modelUuid) {
     return mapSheetLearnRepository.getInferenceAiResultById(id, modelUuid);
   }
 
@@ -364,7 +376,7 @@ public class InferenceResultCoreService {
 
   /**
    * @param compareYear 비교년도
-   * @param targetYear 기준년도
+   * @param targetYear  기준년도
    * @return
    */
   public Integer getLearnStage(Integer compareYear, Integer targetYear) {
diff --git a/src/main/java/com/kamco/cd/kamcoback/postgres/repository/label/LabelAllocateRepositoryImpl.java b/src/main/java/com/kamco/cd/kamcoback/postgres/repository/label/LabelAllocateRepositoryImpl.java
index 685a8217..aa2dc8f8 100644
--- a/src/main/java/com/kamco/cd/kamcoback/postgres/repository/label/LabelAllocateRepositoryImpl.java
+++ b/src/main/java/com/kamco/cd/kamcoback/postgres/repository/label/LabelAllocateRepositoryImpl.java
@@ -34,7 +34,6 @@ import com.querydsl.core.types.dsl.CaseBuilder;
 import com.querydsl.core.types.dsl.Expressions;
 import com.querydsl.core.types.dsl.NumberExpression;
 import com.querydsl.core.types.dsl.StringExpression;
-import com.querydsl.jpa.JPAExpressions;
 import com.querydsl.jpa.impl.JPAQueryFactory;
 import jakarta.persistence.EntityManager;
 import jakarta.persistence.EntityNotFoundException;
diff --git a/src/main/java/com/kamco/cd/kamcoback/scheduler/service/MapSheetInferenceJobService.java b/src/main/java/com/kamco/cd/kamcoback/scheduler/service/MapSheetInferenceJobService.java
index 36ad1059..bbb2d65c 100644
--- a/src/main/java/com/kamco/cd/kamcoback/scheduler/service/MapSheetInferenceJobService.java
+++ b/src/main/java/com/kamco/cd/kamcoback/scheduler/service/MapSheetInferenceJobService.java
@@ -11,7 +11,6 @@ import com.kamco.cd.kamcoback.inference.dto.InferenceProgressDto;
 import com.kamco.cd.kamcoback.inference.dto.InferenceResultDto.SaveInferenceAiDto;
 import com.kamco.cd.kamcoback.inference.dto.InferenceResultDto.Status;
 import com.kamco.cd.kamcoback.inference.dto.InferenceSendDto;
-import com.kamco.cd.kamcoback.inference.dto.InferenceSendDto.pred_requests_areas;
 import com.kamco.cd.kamcoback.postgres.core.InferenceResultCoreService;
 import com.kamco.cd.kamcoback.scheduler.dto.JobStatusDto;
 import jakarta.transaction.Transactional;
@@ -48,137 +47,247 @@ public class MapSheetInferenceJobService {
   @Value("${inference.url}")
   private String inferenceUrl;
 
-  /** 추론 진행 배치 1분 */
+  /**
+   * 추론 진행 배치 1분
+   */
   @Scheduled(fixedDelay = 60_000)
   @Transactional
   public void runBatch() {
-
-    if ("local".equalsIgnoreCase(profile)) {
+    if (isLocalProfile()) {
       return;
     }
 
     try {
-      InferenceBatchSheet batchSheet =
-          inferenceResultCoreService.getInferenceResultByStatus(Status.IN_PROGRESS.getId());
-
-      if (batchSheet == null) {
+      // 진행중 배치 조회
+      InferenceBatchSheet sheet = findInProgressSheet();
+      if (sheet == null) {
         return;
       }
 
-      HttpHeaders headers = new HttpHeaders();
-      headers.setContentType(MediaType.APPLICATION_JSON);
-      headers.setAccept(List.of(MediaType.APPLICATION_JSON));
-
-      Long batchId = 0L;
-
-      if (batchSheet.getM3BatchId() != null) {
-        batchId = batchSheet.getM3BatchId();
-      } else if (batchSheet.getM2BatchId() != null) {
-        batchId = batchSheet.getM2BatchId();
-      } else if (batchSheet.getM1BatchId() != null) {
-        batchId = batchSheet.getM1BatchId();
-      }
-
-      if (batchId == 0L) {
+      // 배치 아이디 가져오기
+      Long batchId = resolveBatchId(sheet);
+      if (batchId == null || batchId == 0L) {
         return;
       }
 
-      String url = batchUrl + "/" + batchId;
-
-      ExternalCallResult<String> result =
-          externalHttpClient.call(url, HttpMethod.GET, null, headers, String.class);
-
-      int status = result.statusCode();
-      if (status < 200 || status >= 300) {
+      // 추론실행 상태 정보 가져오기
+      JobStatusDto job = fetchJobStatus(batchId);
+      if (job == null) {
         return;
       }
 
-      String json = result.body();
-      JobStatusDto dto = objectMapper.readValue(json, JobStatusDto.class);
-
-      int totalJobs = dto.getTotalJobs();
-      int completedJobs = dto.getCompletedJobs();
-      int failedJobs = dto.getFailedJobs();
-
-      // 성공, 실패 값 더해서 total 과 같으면 완료
-      String inferStatus = setStatus(totalJobs, completedJobs, failedJobs);
-
-      if ("COMPLETED".equals(inferStatus)) {
-        String type = batchSheet.getRunningModelType();
-
-        if (type.equals("M1")) {
-          // M1 완료되었으면 M2 실행
-          startInference(
-              batchSheet.getId(), batchSheet.getUuid(), "M2", batchSheet.getM2ModelUuid());
-          // 종료시간
-          updateProcessingEndTimeByModel(batchSheet.getUuid(), ZonedDateTime.now(), "M1");
-        } else if (type.equals("M2")) {
-          // M2 완료되었으면 M3 실행
-          startInference(
-              batchSheet.getId(), batchSheet.getUuid(), "M3", batchSheet.getM3ModelUuid());
-          // 종료시간
-          updateProcessingEndTimeByModel(batchSheet.getUuid(), ZonedDateTime.now(), "M2");
-        } else if (type.equals("M3")) {
-          // 완료
-          SaveInferenceAiDto saveInferenceAiDto = new SaveInferenceAiDto();
-          saveInferenceAiDto.setUuid(batchSheet.getUuid());
-          saveInferenceAiDto.setStatus(Status.END.getId());
-          saveInferenceAiDto.setInferEndDttm(ZonedDateTime.now());
-          saveInferenceAiDto.setType(type);
-          inferenceResultCoreService.update(saveInferenceAiDto);
-          // 종료시간
-          updateProcessingEndTimeByModel(batchSheet.getUuid(), ZonedDateTime.now(), "M3");
-        }
+      if (isCompleted(job)) {
+        // 완료 처리
+        onCompleted(sheet, job);
       } else {
-        SaveInferenceAiDto saveInferenceAiDto = new SaveInferenceAiDto();
-        saveInferenceAiDto.setUuid(batchSheet.getUuid());
-        saveInferenceAiDto.setStatus(Status.IN_PROGRESS.getId());
-        saveInferenceAiDto.setDetectEndCnt((long) (completedJobs + failedJobs));
-        inferenceResultCoreService.update(saveInferenceAiDto);
+        // 진행중 처리
+        onProcessing(sheet, job);
       }
 
     } catch (JsonProcessingException e) {
-      Thread.currentThread().interrupt();
-      log.error("배치 중 인터럽트 발생", e);
+      // JSON 파싱 오류는 interrupt 대상 아님
+      log.error("배치 중 JSON 파싱 오류", e);
+    } catch (Exception e) {
+      log.error("배치 처리 중 예외", e);
     }
   }
 
-  private void startInference(Long id, UUID uuid, String type, UUID modelUuid) {
+  /**
+   * 진행중 배치 조회
+   *
+   * @return
+   */
+  private InferenceBatchSheet findInProgressSheet() {
+    return inferenceResultCoreService.getInferenceResultByStatus(Status.IN_PROGRESS.getId());
+  }
 
-    InferenceProgressDto progressDto =
-        inferenceResultCoreService.getInferenceAiResultById(id, type, modelUuid);
+  /**
+   * batchId 결정
+   *
+   * @param sheet
+   * @return
+   */
+  private Long resolveBatchId(InferenceBatchSheet sheet) {
+    // M3 > M2 > M1
+    if (sheet.getM3BatchId() != null) {
+      return sheet.getM3BatchId();
+    }
+    if (sheet.getM2BatchId() != null) {
+      return sheet.getM2BatchId();
+    }
+    if (sheet.getM1BatchId() != null) {
+      return sheet.getM1BatchId();
+    }
+    return 0L;
+  }
 
-    String inferenceType = "";
+  /**
+   * 추론실행 상태 정보 가져오기
+   *
+   * @param batchId
+   * @return
+   * @throws JsonProcessingException
+   */
+  private JobStatusDto fetchJobStatus(Long batchId) throws JsonProcessingException {
+    String url = batchUrl + "/" + batchId;
 
-    if (type.equals("M1")) {
-      inferenceType = "G1";
-    } else if (type.equals("M2")) {
-      inferenceType = "G2";
-    } else if (type.equals("M3")) {
-      inferenceType = "G3";
+    ExternalCallResult<String> result =
+      externalHttpClient.call(url, HttpMethod.GET, null, jsonHeaders(), String.class);
+
+    int status = result.statusCode();
+    if (status < 200 || status >= 300) {
+      return null;
     }
 
-    pred_requests_areas predRequestsAreas = new pred_requests_areas();
+    return objectMapper.readValue(result.body(), JobStatusDto.class);
+  }
+
+  private HttpHeaders jsonHeaders() {
+    HttpHeaders headers = new HttpHeaders();
+    headers.setContentType(MediaType.APPLICATION_JSON);
+    headers.setAccept(List.of(MediaType.APPLICATION_JSON));
+    return headers;
+  }
+
+  /**
+   * 완료 판단
+   *
+   * @param dto
+   * @return
+   */
+  private boolean isCompleted(JobStatusDto dto) {
+    return dto.getTotalJobs() <= (dto.getCompletedJobs() + dto.getFailedJobs());
+  }
+
+  /**
+   * 완료 처리
+   *
+   * @param sheet
+   * @param job
+   */
+  private void onCompleted(InferenceBatchSheet sheet, JobStatusDto job) {
+    String currentType = sheet.getRunningModelType();
+    ZonedDateTime now = ZonedDateTime.now();
+
+    // 현재 모델 종료 업데이트
+    updateProcessingEndTimeByModel(job, sheet.getUuid(), now, currentType);
+
+    // M3이면 전체 종료
+    if ("M3".equals(currentType)) {
+      endAll(sheet, now);
+      return;
+    }
+
+    // 다음 모델 실행 (M1->M2, M2->M3)
+    String nextType = nextModelType(currentType);
+    UUID nextModelUuid = resolveModelUuid(sheet, nextType);
+
+    // 다음 모델 추론 실행
+    startInference(sheet.getId(), sheet.getUuid(), nextType, nextModelUuid);
+  }
+
+  /**
+   * 추론 종료 할때 update
+   *
+   * @param sheet
+   * @param now
+   */
+  private void endAll(InferenceBatchSheet sheet, ZonedDateTime now) {
+    SaveInferenceAiDto save = new SaveInferenceAiDto();
+    save.setUuid(sheet.getUuid());
+    save.setStatus(Status.END.getId());
+    save.setInferEndDttm(now);
+    save.setType("M3"); // 마지막 모델 기준
+    inferenceResultCoreService.update(save);
+  }
+
+  /**
+   * 다음 실행할 모델 타입 조회
+   *
+   * @param currentType
+   * @return
+   */
+  private String nextModelType(String currentType) {
+    if ("M1".equals(currentType)) {
+      return "M2";
+    }
+    if ("M2".equals(currentType)) {
+      return "M3";
+    }
+    throw new IllegalArgumentException("Unknown runningModelType: " + currentType);
+  }
+
+  /**
+   * 모델 정보 UUID 가져오기
+   *
+   * @param sheet
+   * @param type
+   * @return
+   */
+  private UUID resolveModelUuid(InferenceBatchSheet sheet, String type) {
+    if ("M1".equals(type)) {
+      return sheet.getM1ModelUuid();
+    }
+    if ("M2".equals(type)) {
+      return sheet.getM2ModelUuid();
+    }
+    if ("M3".equals(type)) {
+      return sheet.getM3ModelUuid();
+    }
+    throw new IllegalArgumentException("Unknown type: " + type);
+  }
+
+  /**
+   * 진행중 처리
+   *
+   * @param sheet
+   * @param job
+   */
+  private void onProcessing(InferenceBatchSheet sheet, JobStatusDto job) {
+    SaveInferenceAiDto save = new SaveInferenceAiDto();
+    save.setUuid(sheet.getUuid());
+    save.setStatus(Status.IN_PROGRESS.getId());
+    save.setPendingJobs(job.getPendingJobs());
+    save.setRunningJobs(job.getRunningJobs());
+    save.setCompletedJobs(job.getCompletedJobs());
+    save.setFailedJobs(job.getFailedJobs());
+    inferenceResultCoreService.update(save);
+  }
+
+  /**
+   * 다음 모델 추론 실행
+   *
+   * @param id
+   * @param uuid
+   * @param type
+   * @param modelUuid
+   */
+  private void startInference(Long id, UUID uuid, String type, UUID modelUuid) {
+
+    // 추론 실행 api 파라미터 조회
+    InferenceProgressDto progressDto =
+      inferenceResultCoreService.getInferenceAiResultById(id, modelUuid);
+
+    // ai 에 맞는 모델 명으로 변경
+    String inferenceType = modelToInferenceType(type);
+
+    InferenceSendDto.pred_requests_areas predRequestsAreas = new InferenceSendDto.pred_requests_areas();
     predRequestsAreas.setInput1_year(progressDto.getPred_requests_areas().getInput1_year());
     predRequestsAreas.setInput2_year(progressDto.getPred_requests_areas().getInput2_year());
-    predRequestsAreas.setInput1_scene_path(
-        progressDto.getPred_requests_areas().getInput1_scene_path());
-    predRequestsAreas.setInput2_scene_path(
-        progressDto.getPred_requests_areas().getInput2_scene_path());
+    predRequestsAreas.setInput1_scene_path(progressDto.getPred_requests_areas().getInput1_scene_path());
+    predRequestsAreas.setInput2_scene_path(progressDto.getPred_requests_areas().getInput2_scene_path());
 
     InferenceSendDto m = new InferenceSendDto();
     m.setPred_requests_areas(predRequestsAreas);
     m.setModel_version(progressDto.getModelVersion());
     m.setCd_model_path(progressDto.getCdModelPath() + "/" + progressDto.getCdModelFileName());
-    m.setCd_model_config(
-        progressDto.getCdModelConfigPath() + "/" + progressDto.getCdModelConfigFileName());
-    m.setCls_model_path(
-        progressDto.getCdModelClsPath() + "/" + progressDto.getCdModelClsFileName());
+    m.setCd_model_config(progressDto.getCdModelConfigPath() + "/" + progressDto.getCdModelConfigFileName());
+    m.setCls_model_path(progressDto.getCdModelClsPath() + "/" + progressDto.getCdModelClsFileName());
     m.setCls_model_version(progressDto.getClsModelVersion());
     m.setCd_model_type(inferenceType);
     m.setPriority(progressDto.getPriority());
 
-    // 추론 다음모델 실행
+    // 추론 실행 api 호출
     Long batchId = ensureAccepted(m);
 
     SaveInferenceAiDto saveInferenceAiDto = new SaveInferenceAiDto();
@@ -191,71 +300,111 @@ public class MapSheetInferenceJobService {
     inferenceResultCoreService.update(saveInferenceAiDto);
   }
 
+  /**
+   * ai 에 맞는 모델 명으로 변경
+   *
+   * @param type 모델 타입
+   * @return String
+   */
+  private String modelToInferenceType(String type) {
+    if ("M1".equals(type)) {
+      return "G1";
+    }
+    if ("M2".equals(type)) {
+      return "G2";
+    }
+    if ("M3".equals(type)) {
+      return "G3";
+    }
+    throw new IllegalArgumentException("Unknown type: " + type);
+  }
+
+  /**
+   * api 호출
+   *
+   * @param dto
+   * @return
+   */
   private Long ensureAccepted(InferenceSendDto dto) {
-    log.info("dto null? {}", dto == null);
-    ObjectMapper om = new ObjectMapper();
-    try {
-      log.info("dto json={}", om.writeValueAsString(dto));
-    } catch (Exception e) {
-      log.error(e.getMessage());
+    if (dto == null) {
+      log.warn("not InferenceSendDto dto");
+      throw new CustomApiException("BAD_REQUEST", HttpStatus.BAD_REQUEST);
     }
 
-    HttpHeaders headers = new HttpHeaders();
-    headers.setContentType(MediaType.APPLICATION_JSON);
-    headers.setAccept(List.of(MediaType.APPLICATION_JSON));
+    // 1) 요청 로그
+    try {
+      log.debug("Inference request dto={}", objectMapper.writeValueAsString(dto));
+    } catch (JsonProcessingException e) {
+      log.warn("Failed to serialize inference dto", e);
+    }
 
-    // TODO 추후 삭제
+    // 2) local 환경 임시 처리
     if ("local".equals(profile)) {
       if (dto.getPred_requests_areas() == null) {
-        dto.setPred_requests_areas(new InferenceSendDto.pred_requests_areas());
+        throw new IllegalStateException("pred_requests_areas is null");
       }
-
       dto.getPred_requests_areas().setInput1_scene_path("/kamco-nfs/requests/2023_local.geojson");
       dto.getPred_requests_areas().setInput2_scene_path("/kamco-nfs/requests/2024_local.geojson");
     }
 
+    // 3) HTTP 호출
+    HttpHeaders headers = new HttpHeaders();
+    headers.setContentType(MediaType.APPLICATION_JSON);
+    headers.setAccept(List.of(MediaType.APPLICATION_JSON));
+
     ExternalCallResult<String> result =
-        externalHttpClient.call(inferenceUrl, HttpMethod.POST, dto, headers, String.class);
+      externalHttpClient.call(inferenceUrl, HttpMethod.POST, dto, headers, String.class);
 
-    int status = result.statusCode();
-    String body = result.body();
-
-    if (status < 200 || status >= 300) {
+    if (result.statusCode() < 200 || result.statusCode() >= 300) {
+      log.error("Inference API failed. status={}, body={}", result.statusCode(), result.body());
       throw new CustomApiException("BAD_GATEWAY", HttpStatus.BAD_GATEWAY);
     }
 
-    Long batchId = 0L;
-
+    // 4) 응답 파싱
     try {
       List<Map<String, Object>> list =
-          om.readValue(body, new TypeReference<List<Map<String, Object>>>() {});
+        objectMapper.readValue(result.body(), new TypeReference<>() {
+        });
 
-      Integer batchIdInt = (Integer) list.get(0).get("batch_id");
-      batchId = batchIdInt.longValue();
       if (list.isEmpty()) {
         throw new IllegalStateException("Inference response is empty");
       }
 
-    } catch (Exception e) {
-      log.error(e.getMessage());
-    }
+      Object batchIdObj = list.get(0).get("batch_id");
+      if (batchIdObj == null) {
+        throw new IllegalStateException("batch_id not found in response");
+      }
 
-    return batchId;
+      return Long.valueOf(batchIdObj.toString());
+
+    } catch (Exception e) {
+      log.error("Failed to parse inference response. body={}", result.body(), e);
+      throw new CustomApiException("INVALID_INFERENCE_RESPONSE", HttpStatus.BAD_GATEWAY);
+    }
   }
 
-  private void updateProcessingEndTimeByModel(UUID uuid, ZonedDateTime dateTime, String type) {
+  /**
+   * 실행중인 profile
+   *
+   * @return
+   */
+  private boolean isLocalProfile() {
+    return "local".equalsIgnoreCase(profile);
+  }
+
+  /**
+   * 모델별 추론 종료 update
+   */
+  private void updateProcessingEndTimeByModel(JobStatusDto dto, UUID uuid, ZonedDateTime dateTime, String type) {
     SaveInferenceAiDto saveInferenceAiDto = new SaveInferenceAiDto();
     saveInferenceAiDto.setUuid(uuid);
     saveInferenceAiDto.setUpdateUid(0L);
     saveInferenceAiDto.setModelEndDttm(dateTime);
     saveInferenceAiDto.setType(type);
+    saveInferenceAiDto.setPendingJobs(dto.getPendingJobs());
+    saveInferenceAiDto.setRunningJobs(dto.getRunningJobs());
+    saveInferenceAiDto.setCompletedJobs(dto.getCompletedJobs());
+    saveInferenceAiDto.setFailedJobs(dto.getFailedJobs());
     inferenceResultCoreService.update(saveInferenceAiDto);
   }
-
-  private String setStatus(int totalJobs, int completedJobs, int failedJobs) {
-    if (totalJobs <= (completedJobs + failedJobs)) {
-      return "COMPLETED";
-    }
-    return "PROCESSING";
-  }
 }