Real-Time IoT Risk Monitoring Dashboards for Commercial Insurance India

Traditional commercial insurance risk monitoring in India operates on an annual cycle. An insurer's risk engineer visits the insured's factory or warehouse before the policy renewal, photographs the premises, checks fire suppression equipment, reviews housekeeping standards, and produces a risk improvement report with recommendations. The next visit is typically 12 months later. Between surveys, the insurer has no visibility into how the risk profile of the premises is evolving: whether a new production line was installed that changed the fire hazard category, whether the suppression system fell out of maintenance compliance, whether electrical loading increased beyond safe capacity, or whether a new chemical store was added near an existing ignition source.

This annual-snapshot approach creates a systematic information gap that affects both pricing and loss prevention. The insurer prices the risk on the basis of what it observed at the last survey; the actual risk at any given moment may be materially different. From a loss prevention standpoint, if a sensor had detected the rising temperature in an electrical panel 48 hours before it caused a fire, the fire could have been prevented. With annual surveys, this signal is never captured.

IoT sensor networks change this equation. A factory instrumented with temperature sensors on electrical panels and heat-generating equipment, vibration sensors on rotating machinery, smoke detectors with real-time alert capability, water sensors in areas vulnerable to flooding or pipe leaks, and cameras with computer vision analytics can transmit live readings to a central platform. That platform can alert plant maintenance teams when readings approach abnormal ranges, trigger automated work orders, and provide a continuous audit trail of the facility's operational and safety condition.

For commercial insurance, the relevant development is the emergence of shared IoT dashboards that give both the insured's risk manager and the insurer's underwriting and claims teams access to live sensor data, with different levels of detail and alerting appropriate to each party's role. This shared visibility creates a fundamentally different insurer-insured relationship: the insurer is no longer an annual visitor examining historical conditions but an ongoing participant in the insured's risk management programme. The implications for premium pricing, loss prevention, claims FNOL, and policy terms are being worked out in practice by Tata AIG, Bajaj Allianz, and a small number of other early movers in the Indian market.

The sensor ecosystem in Indian industrial IoT deployments varies by factory size, industry, and the maturity of the insurer's IoT programme. The following sensor categories are the most commonly deployed in factories participating in Indian commercial insurer IoT programmes as of 2026.

Temperature sensors are the most universally deployed category. In electrical panels and switchgear rooms, temperature sensors detect overheating before it reaches the ignition threshold. In furnace and kiln operations, they monitor process temperatures to detect deviations that indicate equipment malfunction. In cold storage and pharmaceutical facilities, they provide continuous temperature logging to document compliance with storage specifications. Most Indian industrial IoT deployments use Type K thermocouples or PT100 RTD sensors for high-temperature process monitoring and DS18B20 digital sensors or MEMS-based wireless sensors for ambient and equipment temperature monitoring. Data is transmitted via LoRaWAN, MQTT over Wi-Fi, or Modbus to a gateway device that aggregates and transmits to the cloud platform.

Vibration sensors are deployed on rotating machinery: motors, pumps, compressors, fans, and gearboxes. Vibration analysis can detect bearing wear, shaft imbalance, misalignment, and loosening of fasteners before they progress to failure. In insurance terms, vibration monitoring is relevant to machinery breakdown coverage: a bearing failure on a critical pump that causes a 72-hour production shutdown is an insurable event under a machinery loss of profit policy, and early detection allows the insured to schedule planned maintenance that avoids the insured event entirely. MEMS accelerometers are the standard sensor for this application, with FFT (Fast Fourier Transform) analysis running either at the edge (on the sensor gateway) or in the cloud platform to decompose the vibration signal into frequency components that indicate specific fault modes.

Smoke detectors and aspirating smoke detection systems provide earlier fire warning than traditional heat-activated detectors. Aspirating smoke detection (ASD) systems, sold in India under brands including Vesda, Orion, and local equivalents, continuously sample air from the protected space through a network of pipes and detect smoke particles at concentrations far below those that would activate a conventional detector. In a factory context, ASD systems can detect smouldering fires 30 to 90 minutes before they produce visible flame, allowing evacuation and suppression activation before significant damage occurs. For insurance purposes, ASD data logged to a cloud platform provides a continuous record that can be used in claims investigation to reconstruct the timeline of a fire.

Water ingress and flood sensors are deployed in basement mechanical rooms, below raised flooring in electrical and data centre spaces, and in facilities near water bodies or with known drainage problems. These sensors detect water presence before it reaches levels that cause equipment damage, and their alert capability gives plant teams time to investigate and intervene. For commercial property insurance, water damage is a significant claim category: the General Insurance Council reported in 2025 that water-related claims accounted for 18% of commercial property claim value in FY2024-25. Early water detection reduces claim frequency and severity.

CCTV with computer vision analytics represents the most data-rich IoT input for insurance purposes. Modern IP cameras paired with computer vision models can perform real-time monitoring for safety compliance: personal protective equipment usage (hardhat, high-visibility vest, safety footwear detection), unauthorised access to restricted zones, fire and smoke detection from video frames, vehicle movement in pedestrian areas, and housekeeping conditions (blocked emergency exits, improperly stored materials near ignition sources). The computer vision outputs a structured alert stream rather than raw video, which is more tractable for integration with the insurance platform and addresses some of the data privacy concerns that arise from raw video sharing with insurers.

A shared IoT dashboard serves three user groups with different information needs, access levels, and action capabilities. Designing appropriate views for each group is essential to the operational effectiveness of the system and to the data governance framework required under IRDAI's guidelines and the DPDP Act 2023.

The underwriter's view is oriented toward risk assessment and portfolio monitoring. The underwriter sees aggregate risk scores for each insured location in their portfolio, derived from sensor data: a location with consistently good compliance (temperature readings within normal bands, vibration signatures indicating healthy equipment, no smoke or water alerts) receives a positive risk score. A location with frequent sensor alerts, evidence of alert suppression or sensor tampering, or extended periods of offline sensors receives a negative score. The underwriter uses these scores to differentiate renewal pricing, to prioritise risk engineering visits, and to identify locations that warrant endorsement or exclusion review. The underwriter does not see raw sensor data in real time; they see processed scores and summary alert histories.

The risk manager's view is the richest and most real-time. The risk manager at the insured's facility sees live sensor readings for every monitored point, alert histories, maintenance action records (was the alert acknowledged and addressed?), and benchmarking against the insurer's portfolio averages for similar facility types. This view is the primary operational tool: it allows the risk manager to detect developing issues, assign maintenance tasks, and demonstrate compliance with the insurer's risk improvement conditions. In facilities with CCTV analytics, the risk manager also sees the compliance monitoring outputs: daily reports on PPE compliance rates, access violation counts, and housekeeping scores.

The claims team's view is activated primarily when a loss event occurs. When a fire, flood, or machinery breakdown is reported, the claims team accesses the sensor data history for the affected location, covering the 72 hours before and after the reported loss date. This data allows the claims assessor to reconstruct the sequence of events: when did the temperature rise begin, was an alarm triggered, was the alarm responded to, was the suppression system activated, and when did conditions return to normal. This reconstruction is directly relevant to three aspects of claims handling: confirming that the reported cause is consistent with the sensor data, assessing whether the insured took reasonable steps to mitigate the loss after the first alert, and quantifying the damage timeline for business interruption calculations. The sensor data also provides a contemporaneous record that is more reliable than insured testimony for claims investigation.

Data access governance

IRDAI's Information Security Guidelines 2023 require that data sharing between insurers and insured entities be governed by documented data sharing agreements that specify the categories of data shared, the purpose, the retention period, the parties' respective security obligations, and the data principal's rights. For IoT data shared between an Indian factory and its insurer, the data sharing agreement must address: whether continuous real-time data flow is permitted or whether the insurer accesses aggregated summaries; how long sensor data is retained and by whom; what happens to the data if the policy is cancelled or not renewed; and the insured's rights to review and correct any risk assessments based on the data. These agreements are now standard in the Tata AIG and Bajaj Allianz IoT programmes, and their structure is informed by guidance from IRDAI's data protection working group, which has been active since 2024.

The premium discount structures attached to IoT compliance programmes in Indian commercial insurance are designed to give insured facilities a financial return on their IoT investment while giving the insurer quantifiable evidence that the risk profile warrants the discount. In 2026, the two most developed programmes in the Indian market are operated by Tata AIG and Bajaj Allianz, with others including HDFC Ergo and New India Assurance at earlier stages.

Tata AIG's industrial IoT programme, which covers fire, engineering, and commercial property policies for medium-to-large manufacturing and warehousing facilities, structures discounts across three compliance tiers. Tier 1, which requires temperature and smoke sensor coverage of all critical areas with 95% uptime and data transmission to the Tata AIG platform, offers a base discount of 5 to 8% on the fire premium. Tier 2, which adds vibration monitoring on all critical rotating equipment and CCTV analytics coverage of all ingress and egress points, offers a total discount of 10 to 12%. Tier 3, the highest tier, requires that the IoT data integration include automatic FNOL triggering (where a sensor alert meeting defined criteria automatically creates a claim intimation in the claims system) and active maintenance workflow integration (where sensor alerts are addressed within defined response time standards), and delivers a total discount of up to 15% on the combined fire and engineering premium.

The Tata AIG discount structure is subject to annual verification. At renewal, the insurer's risk engineering team reviews the sensor uptime records, alert response times, and maintenance action logs for the preceding 12 months. Facilities that maintained Tier 3 compliance throughout the year receive the full Tier 3 discount. Facilities that fell below the required uptime or response time standards for more than two consecutive months are reclassified to a lower tier, with the discount adjusted accordingly. The annual verification creates an ongoing incentive for the insured to maintain IoT infrastructure quality rather than installing sensors for the initial discount and then allowing the system to degrade.

Bajaj Allianz's IoT programme, announced in collaboration with a group of industrial IoT platform providers in Q2 2025, takes a slightly different structure. Rather than a tiered discount, it uses a dynamic premium adjustment mechanism where the premium at renewal is computed partly on the basis of the IoT risk score accumulated over the preceding year. Facilities with consistently good sensor data (few alerts, rapid alert response, sensor uptime above 98%) can achieve discounts of 8 to 13% against the base premium. Facilities with poor performance see no discount or a small loading. The dynamic mechanism has the advantage of providing a continuous incentive rather than a threshold effect, but it requires more sophisticated actuarial infrastructure to implement and to explain to policyholders.

ROI for industrial facilities varies by facility size and insurance premium scale. A manufacturing facility paying INR 1 crore annually in fire and engineering premium can achieve savings of INR 8 to 15 lakh through a Tier 3 discount. The annual cost of operating a comprehensive sensor network in a medium-size factory (20,000 to 50,000 square metres) is approximately INR 5 to 12 lakh, including hardware amortisation, connectivity, platform fees, and maintenance. For facilities at the upper end of the premium scale, the insurance discount alone pays for the IoT infrastructure, with the operational maintenance savings (avoided breakdowns, reduced energy waste from temperature optimisation) providing additional return.

IRDAI has not yet published a standalone regulation specifically governing IoT data in insurance. The applicable regulatory framework is assembled from several existing instruments, with supplementary guidance from IRDAI's data protection working group and from the broader DPDP Act 2023 compliance obligations.

The primary IRDAI instrument is the IRDAI (Protection of Policyholders' Interests) Regulations 2017, which require that policyholders be informed of any conditions attached to their coverage, including conditions requiring data sharing as a condition of a premium discount. When a Tata AIG or Bajaj Allianz IoT discount programme requires the insured to share live sensor data, this requirement must be disclosed in the policy schedule and explained in the proposal process. The insured's consent to data sharing cannot be implied from the act of purchasing the policy; it must be explicit and documented.

The IRDAI Information Security Guidelines 2023 provide the operational security framework. They require that data transmitted from insured premises to the insurer's platform be encrypted in transit and at rest, that access to the data be restricted to authorised personnel on a need-to-know basis, and that the insurer maintain an audit trail of who accessed the data and for what purpose. For real-time IoT data, these requirements translate to: TLS 1.3 encryption on all data transmissions, role-based access controls on the dashboard platform, and access logs that are retained for the period specified in the data retention policy.

The DPDP Act 2023 adds further requirements. Section 6 requires that the consent for data processing be specific to the purpose for which the data is collected: an insured's consent to share temperature sensor data for premium discounting purposes does not automatically extend to sharing that data for product development or for sharing with reinsurers. Each additional purpose requires separate consent. Section 8 requires data minimisation: the insurer may collect only the data necessary for the stated purpose. For underwriting purposes, aggregated sensor scores may be sufficient, and the regulatory position on whether raw sensor data can be retained by the insurer (rather than only the processed risk scores) is still being clarified.

The RBI's Account Aggregator framework, which provides a consent-based data-sharing infrastructure for financial data, does not currently cover IoT data. IRDAI's data protection working group has been in discussions since 2024 about whether a similar consent infrastructure could be built for insurance-relevant IoT data, or whether the existing AA framework could be extended. A pilot is expected in 2026 that would use the AA consent flow to govern insured factories' sharing of IoT data with insurers, providing a standardised and auditable consent mechanism that protects both parties.

Until a dedicated IoT data framework is published, insurers operating IoT discount programmes rely on bespoke data sharing agreements that are negotiated with each insured. The agreements cover the data categories, purpose, retention, security standards, and the insured's right to withdraw data sharing (with the consequence of losing the premium discount). IRDAI has informally indicated that it expects these bespoke agreements to meet the minimum standards set by its Information Security Guidelines and the DPDP Act, and that a standardised template framework is being developed.

One of the most operationally significant applications of IoT data in commercial insurance is automatic first notification of loss (FNOL) triggering. When a sensor detects an event that meets the criteria for an insured loss, the IoT platform can automatically create a claim intimation in the insurer's claims management system, bypassing the need for the insured to initiate the claim manually. This capability compresses the time between loss occurrence and claim registration from hours or days to minutes.

The alert-to-claim pipeline requires careful design to avoid excessive false FNOL submissions. A temperature sensor that regularly reads above normal limits due to seasonal ambient conditions should not trigger a claim intimation each time. The FNOL trigger is therefore not based on a single sensor reading but on a combination of signals: sensor reading above threshold for a defined period, corroborating signals from adjacent sensors (for example, a smoke sensor reading combined with elevated temperature on the same electrical panel), or a sensor reading that exceeds a severity threshold (a sudden step-change in temperature rather than a gradual rise). The alert-to-claim logic is configured jointly by the insurer's technology team and the insured's risk management team during the IoT programme onboarding.

When an FNOL is automatically triggered, the initial claim record in the insurer's claims management system contains a richer set of information than a manual FNOL. It includes the sensor readings that triggered the alert, the time series of readings for the preceding 2 to 4 hours, the location of the alert within the facility (sensor ID mapped to a floor plan), and the alert classification from the IoT platform (fire, temperature anomaly, water ingress, vibration fault). The claims team receiving this record can immediately assess whether the event is likely to result in an insurable loss and can dispatch a surveyor with specific information about the nature and location of the event.

Tata AIG's automatic FNOL capability, which is part of its Tier 3 IoT programme, has been active for approximately 15 months as of April 2026. In that period, the programme has processed 127 automatic FNOLs from participating facilities. Of these, 89 (70%) were confirmed as genuine loss events that resulted in claims. 31 (24%) were false positives that were cancelled after the insured confirmed no actual damage. 7 (6%) were claims that were automatically created and then reclassified after on-site investigation. The false positive rate is higher than in a human-reviewed FNOL process but is considered acceptable given the speed advantage: the 89 confirmed claims were registered an average of 6.3 hours faster than comparable claims from non-IoT facilities in the same portfolio, and the sensor data available at claim registration reduced average surveyor preparation time by 40%.

Integration with business interruption assessment

For insured events that trigger business interruption coverage, IoT data provides a contemporaneous record of when production was disrupted and when it resumed. The sensor data from the manufacturing equipment (vibration sensors offline because machinery is shut down, temperature returning to ambient because furnaces are not operating, energy consumption dropping to standby levels) provides an objective timeline of the interruption period. This timeline is more reliable than the insured's testimony for BI calculation purposes and reduces the potential for disputes about the start and end date of the interruption, which are among the most common sources of BI claim disagreement between insurers and insured in India.