AI scheduling tool cuts aged care rostering delays

Gadali has published findings on an artificial intelligence scheduling system used by aged care provider ECH. The project was carried out with ECH and Microsoft Elevate.

The system, called Schedtris, cut morning rescheduling time by 50% and recovered 15 hours of scheduling capacity each week in ECH's operations. The average time needed to deal with a vacancy fell from about eight minutes to under four minutes, while the disruption window was reduced from two hours to under one hour.

The work addresses a recurring problem in aged care: reassigning care visits at short notice when staff availability changes. That task often requires schedulers to weigh worker skills, location, continuity of care, travel time and availability within a narrow window, particularly early in the morning.

According to the organisations, Schedtris was deployed within ECH's Microsoft Azure environment and connected on a read-only basis to operational data from AlayaCare. The system uses live scheduling, workforce, skills and availability data to help staff assess reassignment options, while leaving the final decision to human schedulers.

ECH said the process had previously depended heavily on individual judgment under pressure. The new framework is intended to make that judgment more consistent and repeatable across the scheduling team.

"Before Schedtris, a disrupted morning meant schedulers at 7am were scrambling under intense time pressure while holding huge amounts of complexity in their heads. Schedtris gives them structure exactly when the pressure peaks. The agent doesn't make the decision for them, but it helps them make better decisions quickly and with more confidence," said Sharon Paulson, Head of Digital Workplace Services at ECH.

The case is outlined in a white paper arguing that the next phase of AI use in organisations will focus less on individual office productivity and more on redesigning operational processes. In that view, the value lies in improving the speed and quality of decisions within workflows that affect cost, service reliability and workforce coordination.

Process shift

The distinction matters in sectors such as aged care, where service delivery depends on frequent operational decisions rather than one-off automated tasks. Last-minute rostering changes can affect not only staffing efficiency but also continuity of care, because clients are more likely to receive support from familiar workers if reassignments are managed quickly.

ECH's use of the system was presented as a model for codifying frontline judgment into a structured process. That approach aims to preserve local knowledge while reducing the risk that important decisions rest solely on the experience of individual staff members.

The organisations also pointed to governance and data controls as part of the deployment. Because the tool sits within ECH's existing Microsoft 365 and Azure setup, it can use live operational data while remaining aligned with existing security and oversight arrangements.

Microsoft Elevate, which worked on the project, framed the initiative as an example of AI being applied to essential public-facing services. It said the focus was on better decision-making in settings where time pressure and service continuity are central.

"Microsoft Elevate is grounded in a simple belief: when we put people first, technology becomes a force multiplier for social impact. Working alongside organisations like ECH, we are exploring how AI can support the delivery of critical human services, enhancing real-time decision-making, strengthening operational resilience and ultimately enabling more responsive and equitable outcomes at scale," said Anita Sood, Microsoft Asia Elevate Commercial Lead at Microsoft.

Sector pressure

The results come as aged care providers face pressure to maintain service levels with limited staffing and rising operational demands. Scheduling is a particularly sensitive area because disruptions can ripple across workers, clients and travel plans within hours.

Gadali said the findings point to a broader lesson for organisations testing AI tools: measurable gains are more likely when projects target a specific business problem, rely on trusted operational data and are paired with changes to the underlying process. The white paper identifies five disciplines behind that approach, including starting with a material business problem, treating data quality as core infrastructure and combining technical work with change management.

For ECH, the practical outcome was a shorter window for reorganising disrupted visits and less manual effort for scheduling staff. The organisations said the system operates in a live environment handling about 20 daily disruptions across ECH's client base of roughly 4,500 people.

The project reflects a growing push by technology suppliers and service providers to demonstrate operational results from AI rather than make broad claims about productivity. In this case, the evidence centres on a narrow but important workflow at the heart of day-to-day aged care delivery: matching the right worker to the right client when plans change at short notice.