Aged care provider ECH has seen morning shift rescheduling times cut in half after deploying Schedtris, an AI-powered workforce scheduling system built on Microsoft Azure. The technology, developed in collaboration with technology firm Gadali and supported by the Microsoft Elevate program, leverages cloud-based machine learning and human-in-the-loop oversight to automate one of healthcare’s most complex logistics puzzles. The result: a 50% reduction in the time it takes to reallocate staff across dozens of care sites each morning, freeing coordinators to focus on higher-value tasks.

Rostering in residential and community aged care is notoriously demanding. Last-minute call-offs, fluctuating patient acuity, and strict regulatory ratios create a daily firefight for schedulers. ECH, which supports thousands of older Australians, manages a workforce spanning personal care workers, nurses, and allied health professionals across multiple locations. Before Schedtris, morning rescheduling could consume up to two hours as coordinators manually juggled spreadsheets and phone calls. Now, the AI engine proposes optimized assignments in minutes, subject to a final human review.

The Engine Behind the Efficiency

Schedtris runs entirely within ECH’s Azure environment, a deliberate choice to keep sensitive workforce and client data within the organization’s security perimeter. The solution uses Azure Kubernetes Service (AKS) to orchestrate containerized microservices, allowing the scheduling engine to scale dynamically during peak morning hours. Under the hood, a combinatorial optimization model trained on three years of historical shift data predicts the likelihood of absences and suggests replacements that minimize travel time, respect contract clauses, and maintain continuity of care.

The AI doesn’t replace the human scheduler. Instead, it acts as a digital assistant, generating a ranked list of roster adjustments. A coordinator reviews the suggestions, makes final decisions, and then deploys the updated roster to the workforce management platform. Gadali, the Melbourne-based AI firm that built Schedtris, described this as “human-in-the-loop by design”—a phrase that has become a guiding principle for AI tools in regulated industries.

How Microsoft Elevate Fast-Tracked Delivery

Gadali participated in the Microsoft Elevate program, an initiative that pairs Microsoft engineering expertise with partners building vertical SaaS solutions. Through Elevate, the Schedtris team gained early access to Azure AI services, architecture reviews, and co-development sprints that compressed the typical delivery timeline. Microsoft Azure Machine Learning handles model training and retraining, while Azure Cognitive Search indexes worker profiles so the system can factor in skills, certifications, and language preferences when matching staff to clients.

The cloud foundation also simplifies compliance. All data remains encrypted at rest and in transit, and Azure Policy ensures that only authorized users can view personally identifiable information. For an aged care provider adhering to the Australian Privacy Principles and the Aged Care Quality Standards, this was non-negotiable.

Inside the 50% Time Savings

Gadali’s report breaks down the time-savings into three components. First, automated absence detection pulls in sick-call notifications from the HR system within seconds, rather than waiting for a phone tree. Second, the matching algorithm evaluates thousands of possible replacements in under 60 seconds, considering factors like proximity to the client’s home, the worker’s existing schedule, and fatigue-management rules. Third, the one-click approval workflow makes it possible to publish changes to the mobile-accessible roster immediately.

During a pilot across 15 ECH residential sites, morning rescheduling dropped from an average of 90 minutes to 45 minutes. Over a month, that reclaims roughly 22 hours of coordinator time—equivalent to half a full-time role. For a sector grappling with burnout and workforce shortages, reclaiming even an hour a day can improve job satisfaction and reduce turnover.

The Human Factor: Trust and Oversight

Despite the efficiency gains, ECH was careful to maintain human judgment at the final decision point. The AI flags recommendations with a confidence score; if confidence drops below a threshold, the system prompts the scheduler to double-check. This guardrail has caught edge cases—such as a worker who recently had a family emergency and shouldn’t be assigned to a particular client—that pure automation would have missed.

Lesley Thompson, ECH’s Director of Workforce Planning, noted in a case study shared by Gadali that “the AI doesn’t replace our team’s expertise; it amplifies it. Schedulers tell me they feel more in control, not less.” The sentiment echoes broader industry acceptance studies where practitioners resist black-box automation but embrace decision-support tools.

Broader Impact on Aged Care Operations

The morning rush isn’t the only beneficiary. Schedtris also handles long-range rostering, generating four-week schedules that balance preference and fairness. Preliminary data suggests a 15% reduction in staff complaints about shift allocations, as the system tracks how often each worker receives undesirable shifts and evens out the load. For clients, improved continuity—keeping the same care worker visiting—has become a measurable quality metric, and early feedback indicates a 10% uplift in client satisfaction scores.

From an IT perspective, the Azure deployment model has proven resilient. During a state-wide fiber outage that knocked out the on-premises legacy scheduler, Schedtris failed over to a secondary Azure region within minutes, ensuring ECH could continue operating. Business continuity in aged care is not just an IT checkbox; it’s a lifeline for vulnerable people who depend on medication visits and personal care at precise times.

The Architecture Behind the Scenes

Schedtris’s technical stack is a blueprint for other regulated industries. The front end is a Progressive Web App (PWA) that works offline and syncs when connectivity returns—critical for coordinators moving between buildings. The backend runs on AKS with Azure Database for PostgreSQL, chosen for its support of JSONB columns that store complex worker availability rules. Azure Logic Apps orchestrate the flow between the employee self-service portal (built on Power Apps) and the scheduling engine.

Model training pipelines use Azure Machine Learning’s automated ML to compare algorithms weekly, ensuring the system adapts to seasonal patterns like flu outbreaks. Retraining happens over the weekend via spot instances, keeping costs low. Gadali estimates that the entire AI infrastructure costs less than $2,000 per month at ECH’s scale, a figure that becomes more compelling as the provider expands.

What This Means for the Windows Ecosystem

While Schedtris runs in the cloud, it’s part of a larger Windows-centric workflow at ECH. Schedulers use Windows 11 devices with Microsoft Edge to access the portal; Active Directory Federation Services (AD FS) handles single sign-on, and Group Policy enforces security baselines. The integration illustrates how modern Azure workloads can coexist with traditional Windows infrastructure, leveraging tools like Windows Admin Center for hybrid management.

Microsoft’s investment in industry clouds, including the Microsoft Cloud for Healthcare, provides a template for similar solutions. Partners like Gadali can plug into Azure Health Data Services, FHIR APIs, and Teams integration to create comprehensive care coordination platforms. Schedtris already sends roster updates to care workers via Microsoft Teams, complete with shift details and client notes.

Looking Ahead: From Scheduling to Predictive Care

Gadali and ECH are exploring next-phase features that move beyond reactive scheduling. One prototype uses Azure AI Anomaly Detector to flag unusual patterns in client behavior—such as missed medications or reduced mobility—that could signal a health decline. Coupled with the roister, this could shift staffing assignments dynamically, sending a registered nurse instead of a personal care worker when the system predicts a higher-acuity day.

Such innovations bring ethical considerations. How much autonomy should an AI have in shifting care levels without human approval? ECH’s governance board is drafting an AI ethics policy, a move Gadali’s CEO, Amir Hossein, says should become standard for any vendor building for the care economy. “We’re not selling an algorithm; we’re selling trust,” he said.

Key Takeaways for IT Leaders

For organizations watching this deployment, several lessons stand out:

  • Start with a narrow, high-pain use case. Morning rescheduling was a contained problem with clear metrics, making ROI easy to prove.
  • Invest in the human-AI interface. Schedtris succeeded because it complemented, rather than replaced, the coordinator’s workflow.
  • Leverage co-engineering programs. The Microsoft Elevate offering shortened the path from concept to production while ensuring architectural best practices.
  • Prioritize data sovereignty from day one. Running the solution in ECH’s own Azure tenant satisfied legal and ethical requirements without lengthy procurement battles.

Aged care is often seen as a laggard in digital transformation, but ECH’s experience shows that targeted AI, deployed thoughtfully, can deliver immediate, measurable benefits. As the sector faces an aging population and chronic staff shortages, tools like Schedtris will move from nice-to-have to essential infrastructure.

While no two organizations are identical, the template—Azure-powered AI, human-in-the-loop design, and rigorous change management—is replicable across healthcare, disability support, and other shift-based industries. The 50% time saving is not just a number; it represents hours returned to coordinators, better care for clients, and a more sustainable workload for frontline managers.