Patient No-Show Prediction and Optimised Appointment Scheduling for Indian Hospitals

Objective: To develop a predictive model and scheduling optimiser that reduces patient no-shows and improves appointment utilisation in Indian healthcare facilities.

No-shows for outpatient appointments are a significant problem in Indian hospitals, leading to resource wastage, increased wait times, and lost revenue. Factors such as traffic, communication gaps, and patient behaviour uniquely impact the Indian context.

The team will analyse historical appointment and attendance data from publicly available Indian hospital datasets (e.g., MIMIC-III, Apollo hospitals' anonymised data, or simulated data if necessary), identify key predictors of no-shows using statistical and machine learning techniques, and design an appointment scheduling algorithm that adapts to predicted attendance probabilities.

Deliverables include: (1) an exploratory analysis of no-show patterns, (2) a predictive model for patient no-shows, (3) a decision-support tool or prototype for scheduling staff that demonstrates the optimiser, and (4) a business case quantifying potential cost savings and improvements in resource utilisation.

The results will help hospital operations managers make data-driven decisions to reduce idle time, improve patient flow, and enhance overall service quality, directly impacting operational efficiency and patient satisfaction.