Exposure watch
IIB Board AI Pack, page 18. First phase. Built here on synthetic data.
Today. Insured property value is reported in monthly and yearly figures. A fast build-up in one district can go unnoticed until a fire or flood shows it.
What would change. Adds up insured property value by district every week, and flags places where it is building up fast. It is the Other Lines vertical's weekly alert.
The typical question
Warehouses in Maharashtra are building up fastest. Their insured value is ₹6,935 cr, up 35.8% on their 13-week average of ₹5,107 cr, so ₹1,827 cr more. One district in the state is flagged as a fast build-up. Leadership sees state totals, so the district is not named here. Next come factories in Maharashtra (₹258 cr added, +2.0%) and factories in Tamil Nadu (₹171 cr added, +2.3%).
| State | Property type | In force | 13-week average | Change | Added | Status |
|---|---|---|---|---|---|---|
| Maharashtra | Warehouses | ₹6,935 cr | ₹5,107 cr | +35.8% | ₹1,827 cr | Fast build-up in a district |
| Maharashtra | Factories | ₹12,939 cr | ₹12,681 cr | +2.0% | ₹258 cr | |
| Tamil Nadu | Factories | ₹7,496 cr | ₹7,325 cr | +2.3% | ₹171 cr | |
| Haryana | Factories | ₹5,389 cr | ₹5,221 cr | +3.2% | ₹168 cr | |
| Andhra Pradesh | Factories | ₹5,141 cr | ₹5,023 cr | +2.4% | ₹118 cr | |
| Telangana | Factories | ₹6,057 cr | ₹5,954 cr | +1.7% | ₹103 cr | |
| Madhya Pradesh | Factories | ₹5,093 cr | ₹5,021 cr | +1.4% | ₹71.9 cr | |
| Gujarat | Factories | ₹8,100 cr | ₹8,034 cr | +0.8% | ₹66.5 cr | |
| Telangana | Warehouses | ₹1,978 cr | ₹1,926 cr | +2.7% | ₹51.6 cr | |
| Delhi | Warehouses | ₹2,206 cr | ₹2,158 cr | +2.2% | ₹48.1 cr |
Worked out from the weekly figures by a fixed rule. No AI was used for this answer. Ranked by insured value added above each place's own 13-week average.
This week's alert: Insured warehouse value is building up fast in one district of Maharashtra
Week ending 4 October 2026. A place is flagged when its insured value in force is at least 10% above its own 13-week average, stands out among districts of a similar size, has risen by at least ₹100 cr, and the rise comes from at least 4 new policies with no single one more than half of it. The full rule is under How it works. This role sees state totals only.
Maharashtra, warehouses: insured value up 35.8% on its 13-week average
Insured warehouse value is building up fast in one district of Maharashtra. Across the state it is up 35.8% on its 13-week average.
How fast, and who wrote it. One district in the state flagged. ₹1,827 cr above its 13-week average. 61 new policies in the last 4 weeks and 105 in 13 weeks, from 24 insurers.
What it could mean. The insured value for warehouses in Maharashtra now makes up 19.0% of the national total for this property type. The new policies are spread across 24 insurers, so only IIB can see the full build-up. One fire, flood or storm where it is building up in the state would now touch more insured value than before.
Written by Azure OpenAI on 7 Oct 2026, 07:41 from the figures below. Every number was checked against them.
Suggested next step. Ask the insurers involved where the new sites are and how they are reinsured, then check again next week. Owner: Other Lines team.
The AI was given the state totals for this build-up only: values, counts and shares. No insurer names or codes. No personal data.
Show the figures the note was written from
{
"state": "Maharashtra",
"property": "warehouses",
"week_ending": "4 October 2026",
"usual_weeks": 13,
"value_in_force_in_state": "₹6,935 cr",
"usual_value_13_week_average": "₹5,107 cr",
"times_usual": 1.36,
"change_on_usual_pct": "35.8%",
"value_added": "₹1,827 cr",
"districts_building_up_fast": 1,
"share_of_national_value_for_this_property_type": "19.0%",
"new_policies_last_4_weeks": 61,
"insurers_with_new_policies_last_13_weeks": 24
}Where insured value sits and moves
States, most value added first
All 36 states and union territories. Choose a row to open its figures, or a state on the map to narrow the list.
| State | In force | Change | Added | Share of India | Status |
|---|---|---|---|---|---|
| Maharashtra | ₹28,324 cr | +8.2% | ₹2,145 cr | 14.1% | Fast build-up in one district |
| Tamil Nadu | ₹13,688 cr | +1.8% | ₹240 cr | 6.8% | |
| Haryana | ₹9,796 cr | +2.3% | ₹220 cr | 4.9% | |
| Telangana | ₹11,483 cr | +1.7% | ₹187 cr | 5.7% | |
| Andhra Pradesh | ₹8,219 cr | +2.0% | ₹160 cr | 4.1% | |
| Delhi | ₹12,691 cr | +1.2% | ₹147 cr | 6.3% | |
| Madhya Pradesh | ₹9,052 cr | +1.2% | ₹108 cr | 4.5% | |
| Karnataka | ₹9,649 cr | +1.0% | ₹97.3 cr | 4.8% | |
| West Bengal | ₹8,636 cr | +1.0% | ₹86.0 cr | 4.3% | |
| Rajasthan | ₹11,023 cr | +0.8% | ₹83.6 cr | 5.5% | |
| Gujarat | ₹14,568 cr | +0.5% | ₹73.9 cr | 7.2% | |
| Kerala | ₹3,884 cr | +1.3% | ₹50.0 cr | 1.9% | |
| Goa | ₹1,101 cr | +4.6% | ₹48.5 cr | 0.5% | |
| Jharkhand | ₹4,937 cr | +0.9% | ₹46.1 cr | 2.5% | |
| Odisha | ₹5,160 cr | +0.4% | ₹19.3 cr | 2.6% | |
| Meghalaya | ₹804 cr | +1.3% | ₹10.2 cr | 0.4% | |
| Nagaland | ₹384 cr | +1.8% | ₹6.8 cr | 0.2% | |
| Tripura | ₹363 cr | +1.1% | ₹3.8 cr | 0.2% | |
| Bihar | ₹6,627 cr | 0.0% | ₹3.1 cr | 3.3% | |
| Andaman and Nicobar Islands | ₹222 cr | +0.7% | ₹1.6 cr | 0.1% | |
| Jammu and Kashmir | ₹893 cr | +0.2% | ₹1.6 cr | 0.4% | |
| Ladakh | ₹254 cr | +0.5% | ₹1.3 cr | 0.1% | |
| Manipur | ₹226 cr | +0.6% | ₹1.2 cr | 0.1% | |
| Arunachal Pradesh | ₹342 cr | +0.2% | ₹85.3 lakh | 0.2% | |
| Lakshadweep | ₹136 cr | +0.4% | ₹57.2 lakh | 0.1% | |
| Himachal Pradesh | ₹303 cr | 0.0% | ₹2.2 lakh | 0.2% | |
| Mizoram | ₹212 cr | 0.0% | -₹1.4 lakh | 0.1% | |
| Sikkim | ₹176 cr | -0.2% | -₹36.7 lakh | 0.1% | |
| Dadra and Nagar Haveli and Daman and Diu | ₹836 cr | -0.1% | -₹42.7 lakh | 0.4% | |
| Puducherry | ₹220 cr | -1.8% | -₹4.0 cr | 0.1% | |
| Chandigarh | ₹966 cr | -0.6% | -₹6.2 cr | 0.5% | |
| Uttarakhand | ₹2,817 cr | -0.2% | -₹6.9 cr | 1.4% | |
| Assam | ₹4,652 cr | -0.1% | -₹6.9 cr | 2.3% | |
| Punjab | ₹5,694 cr | -0.3% | -₹19.7 cr | 2.8% | |
| Uttar Pradesh | ₹18,734 cr | -0.2% | -₹37.5 cr | 9.3% | |
| Chhattisgarh | ₹3,876 cr | -1.1% | -₹43.9 cr | 1.9% |
States on the map
Change shows how far each state's insured value in force is above or below its own 13-week average, in per cent. Insured value shows where the value sits. Darker is higher. Maharashtra is outlined at first because it has a fast build-up this week. The outline then follows the state you choose. Choose a state to open its figures.
Choose a place to see its figures
Insured value in force, week by week
New policies each week
By property type
Insurers behind the new business, last 13 weeks
Insurers shown by code name, never by name, for this role.
Largest new policies, last 13 weeks
How we would measure it
The planted build-up, week by week
The week-by-week replay is district detail. Leadership sees state totals, so this role sees the result above.
Measured live from the weekly figures, insured warehouse value in one district of Maharashtra on 4 October 2026 is 2.50 times its average over the 26 weeks before the build-up. The test data was built to 2.5 times.
Earlier or quieter: the trade-off
The same replay over 91 weeks, the weeks starting 6 January 2025 to 28 September 2026, with only the growth threshold changed.
| Growth at least | Planted build-up flagged | False flags | |
|---|---|---|---|
| 1.05 times | In its first week | 1 | |
| 1.08 times | In its first week | 1 | |
| 1.10 times | 1 week after it began | 0 | This page |
| 1.15 times | 1 week after it began | 0 | |
| 1.20 times | 2 weeks after it began | 0 |
A false flag is a run of flagged weeks anywhere other than the planted build-up, or in it before it began. The replay also set aside 18 jumps that came from one or two large policies, such as one big factory. They raise the value in one place but are not a build-up.
How it works
Every week, the page adds up the sum insured of fire, property and marine cargo policies in force in each district, for each property type. It compares this week with the average of the 13 weeks before, and with districts of a similar size. It flags a place when the value is rising fast, by a material amount, from many new policies. The rule is fixed and explained below. No AI decides what is flagged.
The rule in full, and the technical terms
A district and property type is flagged in a week when all five hold:
- Insured value in force is at least 1.10 times its average over the 13 weeks before.
- That growth scores at least 3.5 on a robust z scale among districts of the same size band, for the same property type, in the same week. A robust z measures how far one district sits from the middle of its group, using the median and the median absolute deviation, so one odd district does not move the yardstick. The size bands are the largest metros, other cities and other districts.
- At least ₹100 cr has been added above the average.
- At least 4 new policies started in the last 4 weeks. Renewals do not count.
- No single new policy in those weeks is more than 50% of the new value. This keeps one large site apart from a build-up.
Concentration is the place's share of all insured value of that property type in India, now and on the 13-week average. The replay runs the same rule over every week that has 13 weeks of history, the weeks starting 6 January 2025 to 28 September 2026, and scores it against the planted build-up. Figures come from the serving view of weekly property exposure. The largest single new policy is read from the curated policy table, without any personal data. The weekly model took 1.4 seconds to build and is kept until the Data store changes.
No personal data is read for this page. A fixed rule flags, not the AI. The AI only drafts the note under an alert, when asked, and a person decides what to do. This use needs no outside data, so nothing on it is mock.