Continuity and Efficiency in Home Care

Introduction

Home care scheduling faces a dual challenge: maximizing continuity of care relationships while operating cost-effectively. Manual scheduling often results in these goals conflicting – fewer unique caregivers per client costs more in travel time, while optimal route planning creates fragmented care relationships.

This research article presents results from an AI-based pilot study of home care scheduling in a Swedish municipality during 2026. By applying scientifically proven methods from Home Health Care Routing and Scheduling Problem (HHCRSP) research, we demonstrate that continuity and efficiency need not be in opposition.

Scientific Background

HHCRSP (Home Health Care Routing and Scheduling Problem) is a complex optimization problem that integrates route planning and scheduling for care personnel. It is an extension of the classical Vehicle Routing Problem (VRP) but with healthcare-specific constraints and objectives.

HHCRSP: Research and Practice

HHCRSP is recognized as a significant variant within the category of Workforce Scheduling and Routing Problems (WSRP) and has been the subject of extensive academic research since the mid-2000s. To optimize continuity, multi-period models are required as the goal of maintaining the same caregiver with the same client necessarily extends over multiple days.

Continuity Metrics

• Number of unique caregivers per client (simplest, but doesn't scale with visit frequency)
• Continuity of Care Index (CCI) – more sophisticated measure accounting for visit distribution
• Kolada N00941 – Swedish standard recommending <11 unique caregivers per client

Methodology

The pilot study was conducted in a Swedish municipality during spring 2026 using an AI-driven scheduling platform based on advanced metaheuristics.

Pilot Study Data

• ~100 home care recipients
• ~4000 planned visits over 2 weeks
• ~40 available caregivers
• 14-day planning horizon

Algorithms Tested

Three different algorithms were tested: From-Request (request-centered, balanced), Minimize Wait Time (field efficiency focus), and Long Solution (extended computation time). The From-Request algorithm dominated the best results.

Key Findings

Continuity: ~7 unique caregivers

Recommended solution achieved 6.92 unique caregivers per client (rounds to ~7), well below the Kolada target of 11 (compared to national average ~15).

Field efficiency: 79%

Field efficiency (visit time/(visit time+travel time)) reached 79.06%, meaning 79% of field time is spent on actual visits.

Assignment: 99%

98.93% of all visits (3,791 of 3,832) could be successfully scheduled, demonstrating practical feasibility.

Compared to the national average of 15 unique caregivers, this represents over 50% improvement in continuity while maintaining high efficiency.

The results clearly show that continuity and efficiency are not a zero-sum game when intelligent AI optimization is applied.

Discussion

The results have practical implications for all stakeholders in home care:

For Municipalities

• Compare travel time against the same source schedule for each scenario
• Systematically achieve Kolada N00941 target (<11 unique caregivers)
• Higher operational efficiency (70%+) means better resource utilization

For Caregivers

• More predictable schedules with fewer ad-hoc changes
• Opportunity to build continuous client relationships

For Clients

• From national average 15 down to 4–7 unique caregivers
• Research shows continuity leads to better health outcomes

References

Conclusion

AI-based optimization of home care scheduling can simultaneously achieve high continuity and high efficiency. The pilot study shows that ~7 unique caregivers per client (well below Kolada target of 11), 79% field efficiency, and 99% assignment rate is achievable. Request-centered algorithm (From-Request) delivers best results.

For municipalities interested in starting a pilot study with AI-optimized scheduling, conducting cost-benefit analysis, or discussing implementation – contact Caire.se.

Continuity and Efficiency in Home Care: AI Optimization