diff --git a/internal/app/service/task/delay_task.go b/internal/app/service/task/delay_task.go
index 058260e..6e57de3 100644
--- a/internal/app/service/task/delay_task.go
+++ b/internal/app/service/task/delay_task.go
@@ -20,6 +20,10 @@ type DelayTask struct {
 	logger        *logs.Logger
 }
 
+func NewDelayTask() Task {
+	return &DelayTask{}
+}
+
 // Execute 执行延迟任务，等待指定的时间
 func (d *DelayTask) Execute() error {
 	d.logger.Infof("任务 %v: 开始延迟 %v...", d.executionTask.TaskID, d.duration)
diff --git a/internal/app/service/task/scheduler.go b/internal/app/service/task/scheduler.go
index caa135e..0208eea 100644
--- a/internal/app/service/task/scheduler.go
+++ b/internal/app/service/task/scheduler.go
@@ -83,7 +83,6 @@ type Scheduler struct {
 	planRepo                repository.PlanRepository
 	analysisPlanTaskManager *AnalysisPlanTaskManager // <--- 注入共享的 Manager
 	progressTracker         *ProgressTracker
-	taskFactory             func(taskType models.TaskType) Task // 调度器需要注入一个任务工厂，用于创建任务实例
 
 	pool     *ants.Pool // 使用 ants 协程池来管理并发
 	wg       sync.WaitGroup
@@ -96,7 +95,6 @@ func NewScheduler(
 	executionLogRepo repository.ExecutionLogRepository,
 	planRepo repository.PlanRepository,
 	analysisPlanTaskManager *AnalysisPlanTaskManager, // <--- 注入 Manager
-	taskFactory func(taskType models.TaskType) Task,
 	logger *logs.Logger,
 	interval time.Duration,
 	numWorkers int) *Scheduler {
@@ -109,7 +107,6 @@ func NewScheduler(
 		pollingInterval:         interval,
 		workers:                 numWorkers,
 		progressTracker:         NewProgressTracker(),
-		taskFactory:             taskFactory,
 		stopChan:                make(chan struct{}), // 初始化停止信号通道
 	}
 }
@@ -266,15 +263,12 @@ func (s *Scheduler) runTask(claimedLog *models.TaskExecutionLog) error {
 		// 执行普通任务
 		task := s.taskFactory(claimedLog.Task.Type)
 		if err := task.ParseParams(s.logger, claimedLog); err != nil {
-
 			s.logger.Errorf("[严重] 任务参数解析失败, 日志ID: %d, 错误: %v", claimedLog.ID, err)
 			return err
 		}
 
 		if err := task.Execute(); err != nil {
-
 			s.logger.Errorf("[严重] 任务执行失败, 日志ID: %d, 错误: %v", claimedLog.ID, err)
-
 			task.OnFailure(err)
 			return err
 		}
@@ -283,6 +277,17 @@ func (s *Scheduler) runTask(claimedLog *models.TaskExecutionLog) error {
 	return nil
 }
 
+// taskFactory 会根据任务类型初始化对应任务
+func (s *Scheduler) taskFactory(taskType models.TaskType) Task {
+	switch taskType {
+	case models.TaskTypeWaiting:
+		return NewDelayTask()
+	default:
+		// TODO 这里直接panic合适吗? 不过这个场景确实不该出现任何异常的任务类型
+		panic("不支持的任务类型")
+	}
+}
+
 // analysisPlan 解析Plan并将解析出的Task列表插入待执行队列中
 func (s *Scheduler) analysisPlan(claimedLog *models.TaskExecutionLog) error {
 	// 创建Plan执行记录
diff --git a/internal/core/application.go b/internal/core/application.go
index 2beddd6..d9af392 100644
--- a/internal/core/application.go
+++ b/internal/core/application.go
@@ -82,7 +82,7 @@ func NewApplication(configPath string) (*Application, error) {
 	analysisPlanTaskManager := task.NewAnalysisPlanTaskManager(planRepo, pendingTaskRepo, executionLogRepo, logger)
 
 	//  初始化任务执行器
-	executor := task.NewScheduler(pendingTaskRepo, executionLogRepo, planRepo, analysisPlanTaskManager, task.TaskFactory, logger, time.Duration(cfg.Task.Interval)*time.Second, cfg.Task.NumWorkers)
+	executor := task.NewScheduler(pendingTaskRepo, executionLogRepo, planRepo, analysisPlanTaskManager, logger, time.Duration(cfg.Task.Interval)*time.Second, cfg.Task.NumWorkers)
 
 	//  初始化 API 服务器
 	apiServer := api.NewAPI(cfg.Server, logger, userRepo, deviceRepo, planRepo, sensorDataRepo, executionLogRepo, tokenService, listenHandler, analysisPlanTaskManager)