1. Why This Exists
India's logistics landscape is not one market — it is 37 overlapping markets with vastly different road quality, monsoon exposure, cold storage capacity, power reliability, and distributor density. A company making a single national distribution decision is implicitly making 37 different risk decisions. A pharmaceutical company routing temperature-sensitive vaccines behaves very differently from an FMCG brand moving ambient packaged goods, but both face the same structural variation across states and UTs.
SENRA scores each state and UT on seven supply chain risk dimensions, normalises them to a common 0–100 scale, and combines them into a composite fragility score. Six sector presets reweight the dimensions to reflect the operational priorities of FMCG, pharma, cold chain, e-commerce, and agriculture supply chains. The result is a decision-support layer that lets planners identify fragile nodes, compare states directly, and stress-test distribution strategies against different risk profiles.
SENRA is designed for logistics planners, FMCG and pharma distribution managers, investors assessing regional exposure, and policy researchers studying infrastructure investment gaps. All underlying data is sourced from Indian government publications and is freely available under India's National Data Sharing and Accessibility Policy.
2. The Seven Dimensions
National highway density as proxy for road quality
Registered distributor/wholesale businesses per capita
Rainfall volume, variability, and flood frequency
Government LEADS index: port, warehouse, service quality
Average annual power outage duration per consumer
Cold storage capacity relative to population
Distributor market spread — fewer dominant players means higher disruption risk
3. Sector Weight Profiles
Six sector presets reweight the seven dimensions to reflect the operational priorities of each industry. All columns sum to 100%.
| Dimension | Default | FMCG | Pharma | Cold Chain | E-Commerce | Agriculture |
|---|---|---|---|---|---|---|
| Road Infrastructure | 22% | 20% | 18% | 15% | 25% | 20% |
| Distributor Density | 18% | 22% | 20% | 12% | 20% | 15% |
| Monsoon Disruption Risk | 18% | 15% | 15% | 20% | 12% | 25% |
| Logistics Access (LEADS) | 16% | 18% | 20% | 18% | 22% | 12% |
| Power Grid Reliability | 12% | 10% | 15% | 20% | 10% | 10% |
| Cold Chain Infrastructure | 8% | 8% | 8% | 12% | 6% | 13% |
| Distributor Concentration | 6% | 7% | 4% | 3% | 5% | 5% |
| Sum | 100% | 100% | 100% | 100% | 100% | 100% |
4. Normalisation Method
Raw dimension values vary in unit and scale — NH km/1000 sq km cannot be directly compared to outage hours per consumer. Each dimension is normalised to a 0–100 subscore using clipped min-max normalisation. Values are first clipped at the 5th and 95th percentile to prevent extreme outliers (e.g. Delhi's business density, Ladakh's road isolation) from compressing all other states into a narrow band. The direction is then inverted for dimensions where a higher raw value means lower risk, so that a high subscore always means high fragility. The final composite score is a weighted sum of all 7 subscores.
The ± uncertainty range shown next to each score reflects the varying proxy quality of each dimension, not statistical sampling error. It is computed as the root-sum-of-squares of each dimension's individual data quality uncertainty, weighted by its contribution to the composite. States with limited published data (Ladakh, Lakshadweep, and most northeastern states) carry a 1.4× penalty on their uncertainty range.
Z-scores were considered but rejected — negative values are harder to communicate to non-technical users. Clipped min-max gives intuitive 0–100 subscores while still handling outliers.
5. Risk Bands
Risk bands are applied to the weighted composite score, not to individual dimension subscores. A state can score CRITICAL on one dimension and LOW overall if other dimensions compensate.
| Band | Score range | Interpretation |
|---|---|---|
| CRITICAL | 70–100 | Severe supply chain fragility — multiple structural vulnerabilities compound risk |
| HIGH | 50–69 | Significant risk — one or two major vulnerabilities require active mitigation |
| MODERATE | 30–49 | Manageable risk — below-average infrastructure in some dimensions |
| LOW | 0–29 | Relatively resilient — strong infrastructure across most dimensions |
6. Limitations
Data vintage. Most underlying data is from 2023–24 government publications. Infrastructure changes — a new NH corridor, a grid upgrade, or an expansion of cold storage capacity — will not be reflected until the next data refresh. SENRA scores represent a snapshot, not a live feed.
Informal economy blind spot. Registered wholesale and distributor businesses (the MCA dataset) do not capture India's substantial informal distribution networks. States with large informal economies — Uttar Pradesh, Bihar — may have their distributor density underestimated, making their scores appear more fragile than ground reality for FMCG distribution specifically.
Monsoon proxy limitations. The IMD composite uses district-level rainfall variability and historical flood frequency. It does not capture drought risk, which is relevant for agriculture supply chains, nor the growing unpredictability of monsoon timing driven by climate change. States vulnerable to dry spells (parts of Maharashtra, Karnataka) may be underscored on monsoon risk.
Concentration proxy. The Distributor Concentration dimension uses a Herfindahl-style concentration score derived from district-level business registrations. It is a proxy for geographic market structure, not a direct measure of supply chain single-point-of-failure risk. A state with one dominant logistics hub (e.g. a major port city) may score high on concentration without that being operationally fragile — the hub may be highly efficient.
Ladakh and Lakshadweep data sparsity. These two UTs have very limited published data. Their scores rely on more estimation than other states and carry higher uncertainty (reflected in the ± range). This will improve as government statistical coverage of new UTs matures.
No dynamic routing. SENRA scores states, not routes. The corridor feature aggregates state-level scores along a geographic path but does not model actual road network topology, live traffic conditions, or real-time disruption events such as strikes, floods, or road closures.