AI Rostering Trial Shows Promising Signs for Wellbeing in Radiographers

TL;DR:

• Real-world pilot study suggests AI-driven self-rostering could boost autonomy, increase fairness, reduce absenteeism, and improve work-life balance for staff
• 71 staff participated across two hospital locations (local and regional)
• At one site, fewer sick days were linked to more fulfilled requests and less work-life conflict.
• While AI showed signs for improving fairness - unmet requests and limited transparency left some staff frustrated.
• The researchers recommend incorporating explainable AI features to make shift allocation decisions clearer and boost user confidence.

AI Self-Scheduling for Radiographers

A recent AI self‑rostering pilot was conducted in two medical imaging departments (covering X-ray, CT, and MRI etc.) within New Zealand. The results signal a positive step towards improving modern workforce rostering and practitioner wellbeing.

The study, led by K. O’Callahan and S. Sitters from Unitec, explored how using RosterLab, an AI-enabled rostering system, could more effectively improve work-life balance, shift-swapping and sick-leave in relation to AI self scheduling while ensuring service demands were met.

With 71 participants, this real-world trial combined surveys, discussion groups and quantitative data on sick leave and shift swaps within a community-based participatory action approach.

Positive Signs For Wellbeing, Reduced Leave, and Autonomy

AI self‑rostering enabled self-rostering style requests for the first time, even while aligning with service requirements.

Staff reported feeling more empowered and autonomous, with a renewed sense of control over their schedules. Some participants perceived boosts in their work-life balance and job satisfaction, which would potentially translate into higher retention and cost savings for the department. Others felt more oversight was needed.

Fairer than Manual, But Human Oversight Still Required

Although the study highlighted clear benefits of AI-driven self-rostering, some participants did express a desire for more visibility into how shift decisions were made and reassurance that the system could be applied equitably in practice.

Unmet requests often left staff frustrated, as no clear reasons were given for why their preferences weren’t met. Some felt worse off than if they hadn’t been given the option to request shifts at all.

This highlights an opportunity to improve transparency through explainable AI (XAI) and ongoing support, which could strengthen trust, fairness, and confidence in the system. The researchers are currently exploring XAI further as a case study.

Why The Study Matters

Traditional self-rostering (often manual and time-consuming) remains a pain point in many healthcare departments, like radiology. Manual approaches can hinder staff autonomy, introduce inequities, and reduce precious clinical time.

This study offers a case for deploying AI-powered rostering alongside human oversight (i.e., the roster maker) to:

• Improve staff wellbeing
• Reinforce a culture of fairness and flexibility
• Reduction in absenteeism

By prioritising fairness, transparency, and user support, departments can harness AI rostering not just as a scheduling tool - but as a strategic lever for workforce resilience and satisfaction.

Want to try AI Self-Rostering?

With clear implementation strategies, transparent communication, and supportive onboarding, departments can successfully introduce AI rostering into their daily practice with tools like RosterLab. Speak with our team about your rostering challenges.