Include unassigned visits in the output for Field Service Routing & other minor improvements (v0.49.0)

Today, we’re announcing version v0.49.0 of the Timefold Platform and v1.7.0 of the Timefold models, available on app.timefold.ai from 2025-08-25.

This release includes several minor UI improvements (including better rounding of distances in output metrics) and bugfixes (including APIs sometimes returning a faulty 500 http error) for the Timefold Platform.

Next to that, this new version of the Timefold Platform comes with updates to these Timefold Models:

Field Service Routing (v1 | Stable)

Unassigned visits in the model output
: We have added a new attribute
unassignedVisits
to the model output. Only the visit IDs are included in the response, not the full visit objects. See Changelog for details. This change makes the finding of unassigned visits easier and more efficient.
Improved scalability of movable visit handling
: The performance of the custom move handling movable visits has been improved, especially for large datasets with many movable visits (assignable to multiple days).
Fixed recommendation vehicle shift dependencies
: Fixed a bug where vehicle shift dependencies were duplicated in the output and where some vehicle shifts weren't considered as dependencies.
Floating breaks can be scheduled during waiting time before travel
: Floating breaks can now be scheduled during the waiting time before travel starts, which allows for more efficient scheduling of breaks. This is especially useful for pinning. See Changelog for details.

Employee Shift Scheduling (v1 | Stable)

Disruption KPI
: When submitting a schedule with already assigned employees, the
disruption
KPI is now always calculated. Before, this Disruption KPI was only calculated when
disruptionRules
were provided. See the Real-time planning feature guide for details.

Please let us know if you have feedback.