AI Attendance Intelligence Center
Attendance health is holding at 91.4%, but 58 students are drifting toward the threshold.
The engine scored every enrolled student against fourteen attendance signals and explained each prediction. Nothing is sent and no case is opened until you approve it.
2.2 points below today, driven by 58 drifting students.
Nightly run finished. 3,842 students scored across 14 signals.
14 students newly entered prediction range.
Route 04 flagged as a shared cause for 5 open cases.
23 students moved from Watch to Good after parent contact.
Attendance Case Queue
Every student the engine placed under observation, ordered by how far the prediction sits below today’s figure.
Liam Walker
Class IV-B · ATN-2041
Falling 4 weeks · Chronic absence forming
Kabir Sethi
Class VIII-A · ATN-1876
Falling 6 weeks · Monday absence pattern
George Kettut
Class II-A · ATN-1655
Falling 3 weeks · Late arrivals compounding
Sophia Reyes
Class III-A · ATN-1420
Falling 2 weeks · Post-holiday drop-off
Aarav Menon
Class VI-B · ATN-1198
Flat 3 weeks · Transport delay exposure
Maria Fernandes
Class IX-C · ATN-1042
Recovering · Medical leave frequency
Why the AI Predicts a Decline
The reasoning chain behind case ATN-2041, in the order the engine followed it. Each link is evidence, not a chart.
Attendance dropped
CriticalPresent days fell from 24 of 25 to 17 of 25 across six weeks. The decline is monotonic, not episodic, which separates it from ordinary illness absence.
Supporting signalsLate arrivals increasing
CriticalArrivals after the bell rose from 2 to 11 per month. In this cohort a late-arrival slope of this size precedes a full absence within four weeks in 8 of 10 cases.
Supporting signalsTransport delays
HighNine of eleven late marks fall on Route 04. The route logged an average 22 minute overrun on the same days, so the cause sits outside the household.
Supporting signalsHoliday return absence
HighThe first two days after every break are absent for this student in four consecutive terms. The pattern is calendar-driven rather than health-driven.
Supporting signalsMedical leave frequency
WatchSix medical leaves were filed this term against a class median of two. Five carry a certificate, so the leave is legitimate but the frequency is itself the signal.
Supporting signalsAssignment absenteeism
WatchAbsences correlate with assignment due dates at 0.71. The student is present for practical periods and absent for submission days, which points to avoidance rather than illness.
Supporting signalsAttendance Recommendation Center
Ranked by expected attendance gained per hour of staff time. Nothing is applied until you approve it.
Notify Parent
Critical prioritySend the predicted-decline letter to 14 households flagged this morning.
Apply recommendationParent Meeting
Critical priorityBook a 20 minute call for the 6 cases where a letter has already gone unanswered.
Apply recommendationTeacher Follow-up
High priorityAsk the class teacher to confirm the cause for 11 students showing avoidance signals.
Apply recommendationRecovery Plan
High priorityOpen a four-week structured plan for the 9 students below the 75% threshold.
Apply recommendationTransport Review
High priorityRoute 04 accounts for 82% of late marks in Block B. Send it for a schedule review.
Apply recommendationMedical Follow-up
Watch prioritySix students filed leave three times above the class median. Refer for a wellbeing check.
Apply recommendationAttendance Reward
Watch priorityRecognise the 23 students who moved from Watch to Good this month.
Apply recommendationStudent Counseling
Watch priorityFour students show an avoidance signature tied to submission days.
Apply recommendationAttendance Recovery Workspace
Where the 58 open cases sit on the recovery track this morning.
31 cases are waiting at Recovery Plan
Plans opened within five days of detection recover 21 points faster than late starts. Eleven of these were detected more than five days ago.
Attendance Copilot
Six things the copilot can draft for you. Everything it produces arrives as a draft for review.
Attendance Intelligence
What the engine noticed this week that the registers do not show on their own.
Fourteen students crossed into prediction range overnight. Eleven share a Monday absence signature, which points at a weekend routine rather than at school conditions.
Twenty-three students gained ground after a parent contact. The median gain was 7 points within 18 days of the first letter.
Nine students have held below 75% for three consecutive months. Each has an open plan; none has had a face-to-face parent meeting yet.
Sixty-four percent of opened plans reached their target. Plans that began within five days of detection outperformed late starts by 21 points.
Three patterns were not visible in the class registers: assignment-day avoidance, first-day-back absence and a Route 04 late cluster.
Six teachers close their loops within 48 hours and their classes recover fastest. Their contact templates are worth copying across Block B.
Route 04 carries 82% of late marks in Block B despite serving 19% of students. A schedule review would lift five separate cases at once.
Eight actions are queued for approval. Applying the top three would cover 31 of the 58 students currently under observation.
Success Center
Where the interventions worked. These are the patterns worth repeating next term.
students held 100% for the full term
gained 7 points or more this month
plans reached their target figure
holds 97.4% across the whole term
cases closed after sustained recovery
Case Audit Timeline
The full history of case ATN-2041, from the first flagged decline to closure. Every AI action is recorded alongside the human decision that followed it.
Attendance Declined
18 Jun 2026 · 09:12
Present days fell below 75% for the first time this term.
AI Prediction Generated
18 Jun 2026 · 09:15
Engine predicted 63% by 18 July at 96% confidence and opened case ATN-2041.
Teacher Assigned
18 Jun 2026 · 11:40
Teena Mathew accepted ownership and confirmed the transport cause.
Parent Notified
19 Jun 2026 · 08:05
Predicted-decline letter delivered and acknowledged by the guardian.
Recovery Started
22 Jun 2026 · 10:00
Four-week plan opened with a transport review and a weekly checkpoint.
Attendance Improved
16 Jul 2026 · 16:30
Attendance recovered to 84%, twelve points above the predicted floor.
Case Closed
29 Jul 2026 · 12:00
Three consecutive stable weeks recorded. Case closed with a monitoring flag.
Resource Center
The templates, rules and documentation behind everything on this page.