You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Fix cross-site data contamination when switching sites. On desktop, a race between persisting the newly selected site and refreshing the dashboards could briefly load the previous site's issues and uncommissioned devices; the site is now persisted before the refresh is triggered. On Android, the previous site's enrichment maps (floors, tenancies, fitting types, control system names) were reused after a switch; they are now cleared.
Settings "Refresh All" now clears caches for the selected site only — other sites' cached data and the global product catalogs are left intact. Cached device test results are likewise deleted per-site.
The Issues heading count now matches the tab counts (distinct affected devices, respecting the integrator filter) instead of counting raw issue rows, so the heading always equals the sum of the tabs.
Rename the Emergency tab to "Predictive Maintenance (Beta)".
v1.0.66
Name exported PDF reports by site, test type, test name, and test start month (e.g. Site-Emergency_Report-function-Monthly_Test-2026-03.pdf) instead of the generic export date. Embedded PDF document titles match.
Summary report filenames use the earliest running instance start date.
New Settings toggle (BETA) to enable the Predictive Maintenance tab, persisted per platform.
Hide the Emergency tab on sites with no emergency devices.
Richer maintenance predictions: predicted date label, expected life, installed years.
Refetch cached duration tests with incomplete results.
Rate limiting: wait a full minute when the per-minute quota is exhausted.
Desktop installer package renamed to "Cloud Connect".
v1.0.20
New Predictive Maintenance tab on the Emergency Testing screen uses linear regression on historical duration test results to predict when each device's battery will fall below the expected pass threshold.
Predictions sorted by urgency: Critical (<3 months), Caution (3-6), Warning (6-12), OK (>12 months).
No additional API calls — predictions computed from existing cached duration test data.