AI-Powered Scheduling & Patient Access Optimization Dashboard

Project cover

Overview

Built an AI-assisted scheduling dashboard in Power BI that identifies hospital bottlenecks, forecasts patient demand, and recommends operational improvements. Analyzed 25K appointment records, used AI visuals for driver analysis, and delivered a decision-ready tool that highlights delay risks, utilization gaps, and actionable optimization strategies.

What I Did

  • Defined the business objective, metric targets, and analysis scope.
  • Built and validated the data, modeling, and reporting workflow.
  • Packaged outputs for stakeholder interpretation and decision support.

Results/Impact

Delivered an analysis workflow with decision-ready outputs and reusable artifacts.

Tech Stack

  • AI, Forecasting, Operational Efficiency, PowerBI

Deliverables

Project Notes

Description: Built an AI-assisted scheduling dashboard in Power BI that identifies hospital bottlenecks, forecasts patient demand, and recommends operational improvements. Analyzed 25K appointment records, used AI visuals for driver analysis, and delivered a decision-ready tool that highlights delay risks, utilization gaps, and actionable optimization strategies. Skills Demonstrated: AI, Forecasting, Operational Efficiency, PowerBI Project Status: Planning

Overview

An interactive Power BI dashboard that uses AI-assisted analytics to identify scheduling bottlenecks, forecast patient demand, and recommend optimization strategies for hospital departments.

Built with Copilot integration, this project simulates how healthcare systems can apply data science and automation to improve patient access and operational efficiency.

Objectives

  • Analyze department-level wait times, no-show rates, and utilization trends.
  • Use AI forecasting to project future appointment volume.
  • Identify key factors driving scheduling delays through Key Influencers analysis.
  • Generate automated Copilot recommendations for operational improvement.

Tools & Skills Used

Category Details
Tools Power BI, Copilot Pro, Excel (for synthetic data prep)
Techniques Data modeling, DAX calculations, Key Influencers & Decomposition Tree, forecasting models
AI Components Power BI Copilot + built-in AI visuals
Data Size 25,000 synthetic appointment records
Key Metrics Wait time (minutes), no-show rate (%), utilization (%), lead time (days)

Dashboard Design

Page 1 - Access Overview Dashboard: Operational Summary

Screenshot 2025-10-30 213821.png

Purpose: Provide a high-level overview of scheduling performance.

Key Features:

  • KPI Cards: Average wait time, no-show rate, and utilization rate.
  • Forecast Chart: 6-month projection of appointment demand with AI forecasting.
  • Bottleneck Analysis: Department-level comparison of wait time and no-show rate.
  • Dynamic Slicers: Filter by department, provider, insurance type, visit type, and telehealth.

Insight Example:

Pediatrics and Cardiology departments show higher no-show rates, while Self-Pay patients experience longer scheduling delays.

Page 2 - AI-Driven Scheduling Insights & Optimization Scenarios

Screenshot 2025-10-30 213839.png

Purpose: Explore -why- delays happen and -how- to improve scheduling efficiency.

Key Features:

  • Key Influencers Visual: Reveals that Self-Pay patients and April appointments increase the likelihood of 15-21-day waits by 13-18%.
  • Decomposition Tree: Explores utilization patterns by Department - Provider - Visit Type.
  • AI Narrative Section: Copilot summarizes insights and suggests process improvements.

Copilot-Generated Example Insight:

-Telehealth appointments reduce average wait times by 1.6 days. Self-Pay segments show 13% higher delay risk. Recommend prioritizing Q2 Telehealth capacity for Pediatrics and automating Self-Pay verification workflows.-

Dataset Summary

image.png

Synthetic dataset generated to simulate realistic hospital scheduling data with:

  • 25,000 rows across Departments, Providers, Insurance types, and Visit types.
  • Variables: Appointment day, slots booked, slots available, wait time (days), no-show flag, utilization ratio, telehealth status, and payer type.
  • Derived Columns: Utilization %, Wait Time Band (for Key Influencers), Year-Month.

Technical Highlights

  • Created calculated columns using DAX:
    • Utilization = SlotsBooked / SlotsAvailable
    • WaitTimeBand for categorical AI analysis.
  • Configured AI Forecast to predict 3-month future booking trends.
  • Leveraged Copilot to produce narrative insights and optimization strategies.

Key Outcomes

Goal Result
Forecast patient demand Achieved 6-month forward projection with confidence interval
Identify bottlenecks Highlighted Self-Pay and Pediatrics as delay drivers
Integrate AI decision support Generated Copilot recommendations for resource reallocation
Create executive-ready visuals Two-page Power BI design with interactive slicers and KPIs

Deliverables

Hospital Operations Efficiency Dashboard.pbix

Hospital_Scheduling_AI_25K.csv

Hospital Efficiency Report.pdf

Reflection

This project pushed me to combine healthcare analytics with AI-assisted business intelligence. Instead of only tracking metrics, I built a dashboard that thinks - so forecasting demand, uncovering patterns, and recommending real actions.

Attribution

Designed and developed by Markuss Saule.