Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Agentic AI for Clinical Urgency Mapping and Queue Optimization in High-Volume Outpatient Departments: A Simulation-Based Evaluation

Ravish Gupta, Saket Kumar, Maulik Dang

Abstract

Outpatient departments (OPDs) in Indian public hospitals face severe overcrowding, with daily volumes reaching 200--8,000 patients~\cite{aiims2020annual}. The prevailing First-Come-First-Served (FCFS) token system treats all patients equally regardless of clinical urgency, leading to dangerous delays for critical cases. We present an agentic AI framework integrating six components: voice-based multilingual symptom capture (modeled), LLM-powered severity prediction, load-aware physician assignment, adaptive queue optimization with urgency drift detection, a multi-objective orchestrator, and a Patient Memory System for longitudinal context-aware triage. Evaluated through discrete-event simulation of a District Hospital in Jabalpur (Madhya Pradesh) with 368 synthetic patients over 30 runs, the framework achieves 94.2\% critical patients seen within 10 minutes (vs.~30.8\% under FCFS), detects $\sim$236 simulated urgency drift events per session (modeled via stochastic deterioration probabilities), identifies $\sim$11.9 additional hidden-critical cases via patient memory, and recomposes queue urgency distribution from 13/36/158/161 (Critical/High/Medium/Low) to $\sim$25/178/115/50 through continuous reassessment, while maintaining comparable throughput ($\sim$40.4 patients/hour).

Agentic AI for Clinical Urgency Mapping and Queue Optimization in High-Volume Outpatient Departments: A Simulation-Based Evaluation

Abstract

Outpatient departments (OPDs) in Indian public hospitals face severe overcrowding, with daily volumes reaching 200--8,000 patients~\cite{aiims2020annual}. The prevailing First-Come-First-Served (FCFS) token system treats all patients equally regardless of clinical urgency, leading to dangerous delays for critical cases. We present an agentic AI framework integrating six components: voice-based multilingual symptom capture (modeled), LLM-powered severity prediction, load-aware physician assignment, adaptive queue optimization with urgency drift detection, a multi-objective orchestrator, and a Patient Memory System for longitudinal context-aware triage. Evaluated through discrete-event simulation of a District Hospital in Jabalpur (Madhya Pradesh) with 368 synthetic patients over 30 runs, the framework achieves 94.2\% critical patients seen within 10 minutes (vs.~30.8\% under FCFS), detects 236 simulated urgency drift events per session (modeled via stochastic deterioration probabilities), identifies 11.9 additional hidden-critical cases via patient memory, and recomposes queue urgency distribution from 13/36/158/161 (Critical/High/Medium/Low) to 25/178/115/50 through continuous reassessment, while maintaining comparable throughput (40.4 patients/hour).

Paper Structure

This paper contains 39 sections, 2 equations, 3 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Contributions
  3. Related Work
  4. OPD Queue Management in India
  5. AI in Healthcare Triage
  6. Agentic AI and LLMs in Healthcare
  7. Gap Analysis
  8. Proposed Framework
  9. Component 1: Voice-Based Symptom Capture (Modeled)
  10. Component 2: LLM-Powered Severity Prediction
  11. Component 3: Load-Aware Physician Assignment
  12. Component 4: Adaptive Queue Optimization
  13. Component 5: Decision-Making Agent (Orchestrator)
  14. Component 6: Patient Memory System
  15. Methodology and Simulation Design
  16. ...and 24 more sections

Figures (3)

  • Figure 1: System architecture of the six-component agentic AI framework. Solid arrows indicate data flow; the dashed arrow represents the 5-minute re-assessment loop for urgency drift detection.
  • Figure 2: Average wait time by urgency level across three strategies (30 runs, error bars show $\pm$1 std). The Agentic AI prioritizes Critical patients at the cost of longer waits for lower-urgency categories. Inset: zoomed Critical-urgency panel with 10-minute clinical target (dashed line).
  • Figure 3: Critical patient wait time distribution (violin plot). The dashed line marks the 10-minute clinical target. FCFS shows wide spread; both triage-aware systems concentrate critical waits near zero.