Action items

Every item comes from an explicit rule over your records: the evidence, the rule that fired, why it matters and the next step. A detected pattern is a reason to investigate, not proof of a specific fault.

Action items

Showing 26 of 26 items. Generated 02 Nov 2026, 08:02 UTC from the stored data.

High priorityConnector healthNew
Persistent deviation from baseline in connector voltage drop
DEMO-03-C1 · Demo Site C — Adyar area (demo)
Next step: Schedule a physical inspection of the connector (contacts, pins, latch, cable strain relief) and repeat the measurement.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 connector voltage drop observations exceeded 35 mV — the larger of the baseline median 32.4 mV plus 3.5 robust spreads (spread 0.648 mV, from the first 12 observations) and +8% of the median. The largest deviation was 22.3 spreads (45%) above baseline; 5 reading(s) were at least 30% above it.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Open the charger →
Workflow
Note (optional, saved with the status change)
Open the alert and its evidence chart
High priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Schedule a physical inspection of the connector (contacts, pins, latch, cable strain relief) and repeat the measurement.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.352 mΩ — the larger of the baseline median 0.314 mΩ plus 3.5 robust spreads (spread 0.011 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 18.4 spreads (62%) above baseline; 5 reading(s) were at least 30% above it.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
High priorityConnector healthNew
Recurring fault evidence in connector-related faults
DEMO-04-C1 · Demo Site D — Velachery area (demo)
Next step: Review the fault codes with the operator and inspect the connector and cable assembly.
Evidence & status ▾Hide ▴
What is happening
15 connector-related faults were logged in the 30 days ending 2026-10-24 (monitor at 3, inspection at 8), versus a historical rate of about 0.4 per 30 days.
Evidence window 24 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Recurring fault evidence” (see Analysis method for thresholds).
Confidence and data quality
Fault codes are logged by the charger and may include causes unrelated to the connector (vehicle-side, network or user handling).
Open the charger →
High priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Inspect the connector and compare against the operator's maintenance limits; consider increasing measurement frequency.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.11 mΩ (43% of the baseline median 0.254 mΩ). 93% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
High priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Inspect the connector and compare against the operator's maintenance limits; consider increasing measurement frequency.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.159 mΩ (51% of the baseline median 0.314 mΩ). 100% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
High priorityConnector healthNew
Connector voltage drop forecast to reach its service limit around 07 Nov 2026
DEMO-03-C1 · Demo Site C — Adyar area (demo)
Next step: Plan a service visit before 03 Nov 2026; record the work on the charger page so EVOLT can check it held.
Likely date07 Nov 202680% range 03 Nov 2026 – 12 Dec 2026
Evidence & status ▾Hide ▴
What is happening
At the recent rate (about 6.18 mV per 30 days, from 12 readings over 77 days), the level would reach the default service limit of 48.6 mV (baseline + 50%; set your own on this charger) around 07 Nov 2026 (80% range: 03 Nov 2026 to 12 Dec 2026).
Why it matters
Servicing before the limit avoids the heating, derating and failed sessions that follow it, and lets the visit be planned rather than urgent.
Rule that triggered it
Theil–Sen rate over the latest readings, Sen 80% interval entirely above zero, earliest date within 30 days.
Open the charger →
High priorityConnector healthNew
Contact resistance is already at its service limit
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Service the connector now, record the work on the charger page, and take a new reading to confirm the fix.
Current level0.492 mΩlimit 0.472 mΩ (default: baseline + 50%)
Evidence & status ▾Hide ▴
What is happening
The current level, 0.492 mΩ (median of the last three readings), is already at or above the default service limit of 0.472 mΩ (baseline + 50%; set your own on this charger).
Why it matters
The connector has passed the level at which it was meant to be serviced; heating and failed sessions become more likely.
Rule that triggered it
Median of the three latest readings ≥ service limit.
Open the charger →
Workflow
Note (optional)
Medium priorityConnector healthNew
Persistent deviation from baseline in connector voltage drop
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 connector voltage drop observations exceeded 36.4 mV — the larger of the baseline median 33.7 mV plus 3.5 robust spreads (spread 0.701 mV, from the first 12 observations) and +8% of the median. The largest deviation was 10.1 spreads (21%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.274 mΩ — the larger of the baseline median 0.254 mΩ plus 3.5 robust spreads (spread 0.006 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 18.7 spreads (42%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-05-C2 · Demo Site E — Anna Nagar area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.232 mΩ — the larger of the baseline median 0.215 mΩ plus 3.5 robust spreads (spread 0.005 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 11.3 spreads (26%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.291 mΩ — the larger of the baseline median 0.26 mΩ plus 3.5 robust spreads (spread 0.009 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 8.4 spreads (29%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Recurring fault evidence in connector-related faults
DEMO-10-C2 · Demo Site J — Vellore central area (demo)
Next step: Review the logged fault codes for a common cause before the next maintenance visit.
Evidence & status ▾Hide ▴
What is happening
3 connector-related faults were logged in the 30 days ending 2026-10-24 (monitor at 3, inspection at 8), versus a historical rate of about 0.2 per 30 days.
Evidence window 24 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Recurring fault evidence” (see Analysis method for thresholds).
Confidence and data quality
Fault codes are logged by the charger and may include causes unrelated to the connector (vehicle-side, network or user handling).
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in connector voltage drop
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), connector voltage drop rose by about 6.99 mV (21% of the baseline median 33.7 mV). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-05-C2 · Demo Site E — Anna Nagar area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.045 mΩ (21% of the baseline median 0.215 mΩ). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.083 mΩ (32% of the baseline median 0.26 mΩ). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Contact resistance forecast to reach its service limit around 18 Nov 2026
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Plan a service visit before 14 Nov 2026; record the work on the charger page so EVOLT can check it held.
Likely date18 Nov 202680% range 14 Nov 2026 – 24 Nov 2026
Evidence & status ▾Hide ▴
What is happening
At the recent rate (about 0.052 mΩ per 30 days, from 10 readings over 63 days since the maintenance on 18 Aug 2026), the level would reach the default service limit of 0.381 mΩ (baseline + 50%; set your own on this charger) around 18 Nov 2026 (80% range: 14 Nov 2026 to 24 Nov 2026).
Why it matters
Servicing before the limit avoids the heating, derating and failed sessions that follow it, and lets the visit be planned rather than urgent.
Rule that triggered it
Theil–Sen rate over the latest readings, Sen 80% interval entirely above zero, earliest date within 30 days.
Open the charger →
Medium priorityData qualityNew
30% of sessions deliver more energy than the vehicle's battery can hold
All stations
Next step: Confirm with the data provider whether charging time is in minutes and whether charging power is an average or a peak value.
Affected sessions302 / 100030.2%
Evidence & status ▾Hide ▴
What is happening
Energy is derived as charging power × charging time (minutes assumed). 302 of 1000 sessions exceed the recorded battery capacity.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Energy, utilisation and revenue figures built on these fields would be overstated.
Rule that triggered it
Derived energy (kWh) > battery capacity (kWh); medium priority when at least 10% of sessions are affected.
Confidence and data quality
The energy figure is derived, not metered; the time unit is not stated in the file.
Workflow
Note (optional)
Medium priorityData qualityNew
Station 1: 6 sessions report more power lost than consumed
Station 1
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions6 / 718.5%
Evidence & status ▾Hide ▴
What is happening
6 of 71 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 11: 3 sessions report more power lost than consumed
Station 11
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 437.0%
Evidence & status ▾Hide ▴
What is happening
3 of 43 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 12: 3 sessions report more power lost than consumed
Station 12
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 525.8%
Evidence & status ▾Hide ▴
What is happening
3 of 52 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 10 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 3: 3 sessions report more power lost than consumed
Station 3
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 595.1%
Evidence & status ▾Hide ▴
What is happening
3 of 59 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 8: 3 sessions report more power lost than consumed
Station 8
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 605.0%
Evidence & status ▾Hide ▴
What is happening
3 of 60 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Low priorityEvidence coverageNew
5 of 86 chargers have too little health data to assess
5 chargers
Next step: Schedule regular contact-resistance or voltage-drop readings for these chargers, then import them.
Not assessable5of 86 chargers
Evidence & status ▾Hide ▴
What is happening
Examples: DEMO-11-C2, DEMO-03-C3, DEMO-07-C2, DEMO-09-C2, DEMO-04-C2.
A charger needs at least 6 baseline readings and 3 recent readings of a primary metric to be assessed.
Why it matters
Problems on these chargers would go unnoticed: no evidence is not the same as no fault.
Rule that triggered it
Health status is “Insufficient data” (missing data is never treated as healthy).
Open the station →
Workflow
Note (optional)
Low priorityEvidence coverageNew
58 chargers are assessed on historical data only
58 chargers
Next step: Import recent measurements for these chargers before making maintenance decisions about them.
Newest reading16 Jul 2026
Evidence & status ▾Hide ▴
What is happening
Their newest reading is more than 45 days old (latest: 16 Jul 2026).
Why it matters
A historical assessment describes the connector at that time, not today.
Rule that triggered it
Newest valid reading older than 45 days → assessment labelled historical.
Workflow
Note (optional)
Low priorityData qualityNew
17 isolated sessions report more power lost than consumed
10 stations
Next step: Ask the data provider how power loss and power consumed are measured, and exclude these sessions from loss reporting until resolved.
Stations affected10
Evidence & status ▾Hide ▴
What is happening
Station 17: 3 of 63; Station 2: 2 of 50; Station 6: 2 of 50; Station 9: 2 of 46; Station 14: 2 of 42; Station 13: 2 of 51; Station 19: 1 of 61; Station 4: 1 of 53; Station 10: 1 of 43; Station 16: 1 of 45.
Why it matters
Individually small, but every such record is a measurement that cannot be correct.
Rule that triggered it
Stations where fewer than 5% of sessions have power_loss > power_consumed.
Workflow
Note (optional)
Low priorityData qualityNew
Charging types do not differ in delivered power
All stations
Next step: Check the charger-type labels against each charger's nameplate, and check for power limits on chargers labelled Fast.
Average power by type26–26.9 kW
Evidence & status ▾Hide ▴
What is happening
V2G: 26.9 kW average over 317 sessions; Conventional: 26 kW average over 331 sessions; Fast: 26.6 kW average over 352 sessions.
Why it matters
If “Fast” chargers deliver the same power as others, either the labels are wrong or fast chargers are being throttled — both matter for planning and for drivers.
Rule that triggered it
Largest gap between charging types' average power is under 10%.
Workflow
Note (optional)

How priority is set

High priority: A connector-health analysis currently recommends inspection (persistent deviation, sustained trend or recurring faults), or a connector is forecast to reach its service limit within 14 days (or is already above it).
Medium priority: A connector forecast to reach its service limit within 30 days, a connector-health pattern to monitor, a station whose high-loss or poor-stability share is statistically above the fleet, or inconsistent records in at least 5 % of a station's sessions.
Low priority: Data gaps, historical-only evidence, low demand relative to other stations, isolated inconsistent records, or an alert whose pattern is no longer detected.

“Marked resolved” records a person’s decision in EVOLT. It does not verify that a charger was repaired: if the data still shows the pattern, the item stays visible under “Marked resolved” and is flagged when its evidence changes.

Showing 26 of 26 items. Generated 02 Nov 2026, 08:02 UTC from the stored data.

High priorityConnector healthNew
Persistent deviation from baseline in connector voltage drop
DEMO-03-C1 · Demo Site C — Adyar area (demo)
Next step: Schedule a physical inspection of the connector (contacts, pins, latch, cable strain relief) and repeat the measurement.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 connector voltage drop observations exceeded 35 mV — the larger of the baseline median 32.4 mV plus 3.5 robust spreads (spread 0.648 mV, from the first 12 observations) and +8% of the median. The largest deviation was 22.3 spreads (45%) above baseline; 5 reading(s) were at least 30% above it.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Open the charger →
Workflow
Note (optional, saved with the status change)
Open the alert and its evidence chart
High priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Schedule a physical inspection of the connector (contacts, pins, latch, cable strain relief) and repeat the measurement.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.352 mΩ — the larger of the baseline median 0.314 mΩ plus 3.5 robust spreads (spread 0.011 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 18.4 spreads (62%) above baseline; 5 reading(s) were at least 30% above it.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
High priorityConnector healthNew
Recurring fault evidence in connector-related faults
DEMO-04-C1 · Demo Site D — Velachery area (demo)
Next step: Review the fault codes with the operator and inspect the connector and cable assembly.
Evidence & status ▾Hide ▴
What is happening
15 connector-related faults were logged in the 30 days ending 2026-10-24 (monitor at 3, inspection at 8), versus a historical rate of about 0.4 per 30 days.
Evidence window 24 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Recurring fault evidence” (see Analysis method for thresholds).
Confidence and data quality
Fault codes are logged by the charger and may include causes unrelated to the connector (vehicle-side, network or user handling).
Open the charger →
High priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Inspect the connector and compare against the operator's maintenance limits; consider increasing measurement frequency.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.11 mΩ (43% of the baseline median 0.254 mΩ). 93% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
High priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Inspect the connector and compare against the operator's maintenance limits; consider increasing measurement frequency.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.159 mΩ (51% of the baseline median 0.314 mΩ). 100% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
Rising resistance, voltage drop or repeated faults in the charging path can lead to heating and failed sessions if left unchecked.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
High priorityConnector healthNew
Connector voltage drop forecast to reach its service limit around 07 Nov 2026
DEMO-03-C1 · Demo Site C — Adyar area (demo)
Next step: Plan a service visit before 03 Nov 2026; record the work on the charger page so EVOLT can check it held.
Likely date07 Nov 202680% range 03 Nov 2026 – 12 Dec 2026
Evidence & status ▾Hide ▴
What is happening
At the recent rate (about 6.18 mV per 30 days, from 12 readings over 77 days), the level would reach the default service limit of 48.6 mV (baseline + 50%; set your own on this charger) around 07 Nov 2026 (80% range: 03 Nov 2026 to 12 Dec 2026).
Why it matters
Servicing before the limit avoids the heating, derating and failed sessions that follow it, and lets the visit be planned rather than urgent.
Rule that triggered it
Theil–Sen rate over the latest readings, Sen 80% interval entirely above zero, earliest date within 30 days.
Open the charger →
High priorityConnector healthNew
Contact resistance is already at its service limit
DEMO-01-C2 · Demo Site A — Guindy area (demo)
Next step: Service the connector now, record the work on the charger page, and take a new reading to confirm the fix.
Current level0.492 mΩlimit 0.472 mΩ (default: baseline + 50%)
Evidence & status ▾Hide ▴
What is happening
The current level, 0.492 mΩ (median of the last three readings), is already at or above the default service limit of 0.472 mΩ (baseline + 50%; set your own on this charger).
Why it matters
The connector has passed the level at which it was meant to be serviced; heating and failed sessions become more likely.
Rule that triggered it
Median of the three latest readings ≥ service limit.
Open the charger →
Workflow
Note (optional)
Medium priorityConnector healthNew
Persistent deviation from baseline in connector voltage drop
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 connector voltage drop observations exceeded 36.4 mV — the larger of the baseline median 33.7 mV plus 3.5 robust spreads (spread 0.701 mV, from the first 12 observations) and +8% of the median. The largest deviation was 10.1 spreads (21%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.274 mΩ — the larger of the baseline median 0.254 mΩ plus 3.5 robust spreads (spread 0.006 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 18.7 spreads (42%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-05-C2 · Demo Site E — Anna Nagar area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.232 mΩ — the larger of the baseline median 0.215 mΩ plus 3.5 robust spreads (spread 0.005 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 11.3 spreads (26%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Persistent deviation from baseline in contact resistance
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Keep monitoring; confirm the next measurement uses the same method and conditions.
Evidence & status ▾Hide ▴
What is happening
5 of the last 5 contact resistance observations exceeded 0.291 mΩ — the larger of the baseline median 0.26 mΩ plus 3.5 robust spreads (spread 0.009 mΩ, from the first 12 observations) and +8% of the median. The largest deviation was 8.4 spreads (29%) above baseline.
Evidence window 26 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Persistent deviation from baseline” (see Analysis method for thresholds).
Confidence and data quality
Based on 26 valid observations. A deviation from this charger's own history is not proof of physical degradation; measurement method changes, sensor faults or operating conditions can produce the same pattern.
Medium priorityConnector healthNew
Recurring fault evidence in connector-related faults
DEMO-10-C2 · Demo Site J — Vellore central area (demo)
Next step: Review the logged fault codes for a common cause before the next maintenance visit.
Evidence & status ▾Hide ▴
What is happening
3 connector-related faults were logged in the 30 days ending 2026-10-24 (monitor at 3, inspection at 8), versus a historical rate of about 0.2 per 30 days.
Evidence window 24 Sept 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Recurring fault evidence” (see Analysis method for thresholds).
Confidence and data quality
Fault codes are logged by the charger and may include causes unrelated to the connector (vehicle-side, network or user handling).
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in connector voltage drop
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), connector voltage drop rose by about 6.99 mV (21% of the baseline median 33.7 mV). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-05-C2 · Demo Site E — Anna Nagar area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.045 mΩ (21% of the baseline median 0.215 mΩ). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Sustained adverse trend in contact resistance
DEMO-06-C1 · Demo Site F — OMR / Sholinganallur area (demo)
Next step: Increase measurement frequency to confirm or rule out the trend.
Evidence & status ▾Hide ▴
What is happening
Over the last 10 observations (63 days), contact resistance rose by about 0.083 mΩ (32% of the baseline median 0.26 mΩ). 91% of point-to-point comparisons were increases (threshold 75%).
Evidence window 22 Aug 2026 – 24 Oct 2026; seen in 1 analysis run(s).
The latest analysis still finds this pattern.
Why it matters
The pattern is not yet strong enough to call for inspection, but it is the kind that can precede one.
Rule that triggered it
Connector-health rule “Sustained adverse trend” (see Analysis method for thresholds).
Confidence and data quality
Trend estimated with the Theil–Sen median slope, which tolerates a minority of outliers. 10 points is a short series; the trend may not continue.
Open the charger →
Medium priorityConnector healthNew
Contact resistance forecast to reach its service limit around 18 Nov 2026
DEMO-13-C2 · Demo Site M — Kilpauk area (demo)
Next step: Plan a service visit before 14 Nov 2026; record the work on the charger page so EVOLT can check it held.
Likely date18 Nov 202680% range 14 Nov 2026 – 24 Nov 2026
Evidence & status ▾Hide ▴
What is happening
At the recent rate (about 0.052 mΩ per 30 days, from 10 readings over 63 days since the maintenance on 18 Aug 2026), the level would reach the default service limit of 0.381 mΩ (baseline + 50%; set your own on this charger) around 18 Nov 2026 (80% range: 14 Nov 2026 to 24 Nov 2026).
Why it matters
Servicing before the limit avoids the heating, derating and failed sessions that follow it, and lets the visit be planned rather than urgent.
Rule that triggered it
Theil–Sen rate over the latest readings, Sen 80% interval entirely above zero, earliest date within 30 days.
Open the charger →
Medium priorityData qualityNew
30% of sessions deliver more energy than the vehicle's battery can hold
All stations
Next step: Confirm with the data provider whether charging time is in minutes and whether charging power is an average or a peak value.
Affected sessions302 / 100030.2%
Evidence & status ▾Hide ▴
What is happening
Energy is derived as charging power × charging time (minutes assumed). 302 of 1000 sessions exceed the recorded battery capacity.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Energy, utilisation and revenue figures built on these fields would be overstated.
Rule that triggered it
Derived energy (kWh) > battery capacity (kWh); medium priority when at least 10% of sessions are affected.
Confidence and data quality
The energy figure is derived, not metered; the time unit is not stated in the file.
Workflow
Note (optional)
Medium priorityData qualityNew
Station 1: 6 sessions report more power lost than consumed
Station 1
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions6 / 718.5%
Evidence & status ▾Hide ▴
What is happening
6 of 71 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 11: 3 sessions report more power lost than consumed
Station 11
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 437.0%
Evidence & status ▾Hide ▴
What is happening
3 of 43 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 12: 3 sessions report more power lost than consumed
Station 12
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 525.8%
Evidence & status ▾Hide ▴
What is happening
3 of 52 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 10 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 3: 3 sessions report more power lost than consumed
Station 3
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 595.1%
Evidence & status ▾Hide ▴
What is happening
3 of 59 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Medium priorityData qualityNew
Station 8: 3 sessions report more power lost than consumed
Station 8
Next step: Check meter calibration and the data export for this station; correct or exclude the affected sessions and re-import.
Inconsistent sessions3 / 605.0%
Evidence & status ▾Hide ▴
What is happening
3 of 60 sessions have power loss greater than power consumed, which cannot happen in a correct measurement.
Period: 01 Jan 2024 – 11 Jan 2024.
Why it matters
Loss figures from this station cannot be trusted until the cause is found, so loss-based findings here may be wrong.
Rule that triggered it
Sessions with power_loss > power_consumed make up at least 5% of the station's sessions.
Open the station →
Workflow
Note (optional)
Low priorityEvidence coverageNew
5 of 86 chargers have too little health data to assess
5 chargers
Next step: Schedule regular contact-resistance or voltage-drop readings for these chargers, then import them.
Not assessable5of 86 chargers
Evidence & status ▾Hide ▴
What is happening
Examples: DEMO-11-C2, DEMO-03-C3, DEMO-07-C2, DEMO-09-C2, DEMO-04-C2.
A charger needs at least 6 baseline readings and 3 recent readings of a primary metric to be assessed.
Why it matters
Problems on these chargers would go unnoticed: no evidence is not the same as no fault.
Rule that triggered it
Health status is “Insufficient data” (missing data is never treated as healthy).
Open the station →
Workflow
Note (optional)
Low priorityEvidence coverageNew
58 chargers are assessed on historical data only
58 chargers
Next step: Import recent measurements for these chargers before making maintenance decisions about them.
Newest reading16 Jul 2026
Evidence & status ▾Hide ▴
What is happening
Their newest reading is more than 45 days old (latest: 16 Jul 2026).
Why it matters
A historical assessment describes the connector at that time, not today.
Rule that triggered it
Newest valid reading older than 45 days → assessment labelled historical.
Workflow
Note (optional)
Low priorityData qualityNew
17 isolated sessions report more power lost than consumed
10 stations
Next step: Ask the data provider how power loss and power consumed are measured, and exclude these sessions from loss reporting until resolved.
Stations affected10
Evidence & status ▾Hide ▴
What is happening
Station 17: 3 of 63; Station 2: 2 of 50; Station 6: 2 of 50; Station 9: 2 of 46; Station 14: 2 of 42; Station 13: 2 of 51; Station 19: 1 of 61; Station 4: 1 of 53; Station 10: 1 of 43; Station 16: 1 of 45.
Why it matters
Individually small, but every such record is a measurement that cannot be correct.
Rule that triggered it
Stations where fewer than 5% of sessions have power_loss > power_consumed.
Workflow
Note (optional)
Low priorityData qualityNew
Charging types do not differ in delivered power
All stations
Next step: Check the charger-type labels against each charger's nameplate, and check for power limits on chargers labelled Fast.
Average power by type26–26.9 kW
Evidence & status ▾Hide ▴
What is happening
V2G: 26.9 kW average over 317 sessions; Conventional: 26 kW average over 331 sessions; Fast: 26.6 kW average over 352 sessions.
Why it matters
If “Fast” chargers deliver the same power as others, either the labels are wrong or fast chargers are being throttled — both matter for planning and for drivers.
Rule that triggered it
Largest gap between charging types' average power is under 10%.
Workflow
Note (optional)

How priority is set

High priority: A connector-health analysis currently recommends inspection (persistent deviation, sustained trend or recurring faults), or a connector is forecast to reach its service limit within 14 days (or is already above it).
Medium priority: A connector forecast to reach its service limit within 30 days, a connector-health pattern to monitor, a station whose high-loss or poor-stability share is statistically above the fleet, or inconsistent records in at least 5 % of a station's sessions.
Low priority: Data gaps, historical-only evidence, low demand relative to other stations, isolated inconsistent records, or an alert whose pattern is no longer detected.

Action items

Action items

Persistent deviation from baseline in connector voltage drop

Persistent deviation from baseline in contact resistance

Recurring fault evidence in connector-related faults

Sustained adverse trend in contact resistance

Sustained adverse trend in contact resistance

Connector voltage drop forecast to reach its service limit around 07 Nov 2026

Contact resistance is already at its service limit

Persistent deviation from baseline in connector voltage drop

Persistent deviation from baseline in contact resistance

Persistent deviation from baseline in contact resistance

Persistent deviation from baseline in contact resistance

Recurring fault evidence in connector-related faults

Sustained adverse trend in connector voltage drop

Sustained adverse trend in contact resistance

Sustained adverse trend in contact resistance

Contact resistance forecast to reach its service limit around 18 Nov 2026

30% of sessions deliver more energy than the vehicle's battery can hold

Station 1: 6 sessions report more power lost than consumed

Station 11: 3 sessions report more power lost than consumed

Station 12: 3 sessions report more power lost than consumed

Station 3: 3 sessions report more power lost than consumed

Station 8: 3 sessions report more power lost than consumed

5 of 86 chargers have too little health data to assess

58 chargers are assessed on historical data only

17 isolated sessions report more power lost than consumed

Charging types do not differ in delivered power

Persistent deviation from baseline in connector voltage drop

Persistent deviation from baseline in contact resistance

Recurring fault evidence in connector-related faults

Sustained adverse trend in contact resistance

Sustained adverse trend in contact resistance

Connector voltage drop forecast to reach its service limit around 07 Nov 2026

Contact resistance is already at its service limit

Persistent deviation from baseline in connector voltage drop

Persistent deviation from baseline in contact resistance

Persistent deviation from baseline in contact resistance

Persistent deviation from baseline in contact resistance

Recurring fault evidence in connector-related faults

Sustained adverse trend in connector voltage drop

Sustained adverse trend in contact resistance

Sustained adverse trend in contact resistance

Contact resistance forecast to reach its service limit around 18 Nov 2026

30% of sessions deliver more energy than the vehicle's battery can hold

Station 1: 6 sessions report more power lost than consumed

Station 11: 3 sessions report more power lost than consumed

Station 12: 3 sessions report more power lost than consumed

Station 3: 3 sessions report more power lost than consumed

Station 8: 3 sessions report more power lost than consumed

5 of 86 chargers have too little health data to assess

58 chargers are assessed on historical data only

17 isolated sessions report more power lost than consumed

Charging types do not differ in delivered power