Hossein Hashemi Doulabi's research team at Concordia University in Montreal has built an artificial intelligence scheduling tool that could transform how hospitals manage their operating rooms — cutting wait times, absorbing emergency surgeries, and keeping costs under control. The breakthrough lies not in raw computing power, but in elegance: their model uses far fewer variables than the widely used previous approach, making it fast enough and practical enough to actually work in real hospitals juggling dozens or even hundreds of operations each week.
Hospitals face a perpetual juggling act. They must decide which operating rooms to open each day, when to start each surgery, and which cases might need to be delayed — all while never knowing when a true emergency will arrive. Get it wrong and patients wait longer than necessary; operating rooms sit idle, eating up costs; or staff burn out from constant crisis management. The challenge compounds when emergency cases arrive mid-week and demand immediate attention, forcing hospitals to either bump scheduled patients or scramble for resources they may not have.
The Concordia team's solution uses reinforcement learning, a type of artificial intelligence that learns through trial and error, to replan the surgical schedule day by day. When an emergency arrives or a patient's condition suddenly worsens, the system can slot them in while minimizing disruptions to the original plan. Rather than chaos, the tool deploys carefully calibrated responses: limited overtime, opening extra rooms when feasible, or deferring a small number of elective cases. In tests using both simulated data and real hospital schedules from Naples, Italy, the system absorbed same-day emergency arrivals with only modest changes to the original plan.
The implications ripple outward. Better scheduling means shorter surgical wait lists — a persistent pain point in healthcare systems worldwide. It reduces day-of-surgery cancellations, which are both traumatic for patients and wasteful for hospitals. And by optimizing room usage and staffing decisions, the tool contributes to better cost control, a concern every healthcare administrator knows intimately. For patients anxious about their surgery date, for surgeons managing their caseload, and for hospital administrators watching budgets, this kind of efficiency matters.
The research, led by Hashemi Doulabi in Concordia's Department of Mechanical, Industrial and Aerospace Engineering, was published in the International Journal of Production Research in 2026 alongside co-author Mahdi Dolatkhah. The team's approach differs fundamentally from brute-force computational methods that bog hospitals down with complexity; instead, they engineered a tool that hospitals can actually use, in real time, when the stakes are highest.
What makes this work particularly hopeful is its practicality. The best scientific breakthrough means nothing if it gathers dust in a lab. By designing their model to handle the messy reality of hospital operations — last-minute emergencies, staff availability, room constraints — the Concordia team has created something that could genuinely ease the burden on overextended healthcare systems. As surgical wait lists grow longer and emergency departments grow more strained, tools that help hospitals work smarter, not just harder, may be exactly what patients and staff need.