# UAE Construction Safety Benchmark — Monthly Worker-Hour Reporting Pack

**Template version:** 1.0  
**For:** Appointed site data owners participating in the UAE Construction Safety Benchmark  
**Reporting frequency:** Monthly; submit by the 10th business day of the following month

## Purpose

This pack collects the **worker-hour denominator** used to calculate comparable construction safety rates. Each participating project submits **one controlled site-level ledger** for the same worker population whose incident records are included in the benchmark. The aim is to reduce reporting burden without sacrificing credibility.

> **Worker-hours are hours actually worked by the defined covered population at the project site during the reporting month.** The same people must be in both the worker-hour denominator and the incident-reporting population.

The benchmark is an opt-in panel, not an official UAE construction census. Do not report an estimate as an actual source value. Do not submit worker names, employee numbers, contact details, identification documents, payroll files, photographs, witness statements, or free-text incident narratives in this pack.

## Forms in this pack

| File | Who completes it | Frequency | Purpose |
|---|---|---|---|
| `01-site-coverage-profile.md` | Site data owner with project HSE lead | At enrolment; update before or immediately after a material scope change | Defines exactly who is included, excluded, and responsible. |
| `02-monthly-worker-hour-register.csv` | Site data owner | Monthly | The mandatory authoritative ledger. One row per non-overlapping workforce group and hour source. |
| `03-subcontractor-worker-hours-annex.csv` | Each subcontractor to main-contractor data owner | Monthly when the subcontractor is in scope | Certified source return that feeds the site ledger. The benchmark receives the consolidated site ledger, not duplicate submissions. |
| `04-monthly-attestation-and-reconciliation.md` | Site data owner and HSE data certifier | Monthly | Confirms totals, incident coverage, data sources, exceptions, and authorised sign-off. |
| `05-data-quality-exception-log.csv` | Site data owner | Only when there is a gap, estimate, late correction, or scope issue | Makes limitations transparent and determines whether a month is eligible for primary benchmark rates. |
| `06-data-dictionary-and-completion-rules.md` | Site data owner and benchmark administrator | Reference | Defines every field, source hierarchy, calculation, eligibility rule, and privacy restriction. |

## Submission model: one site, one ledger, no double counting

The main contractor or another formally appointed **site data owner** submits one consolidated register for each benchmark site and month. Subcontractors should return the annex to that owner. They must **not** submit the same worker-hours separately to the benchmark.

Each row in the main register must represent a mutually exclusive workforce group. If a worker’s hours appear in a direct-labour row, they must not also appear in a labour-supply or subcontractor row. If a worker performs several activities during the month, their hours still appear **once only** in the workforce group that employs or supplies them. Activity is collected on the incident form, not by allocating worker-hours across overlapping trades.

## Required monthly workflow

1. Export the actual monthly hours from the preferred source system for each workforce group.
2. Obtain signed or digitally certified subcontractor annexes where required.
3. Reconcile the workforce groups with the incident-reporting population for the same month.
4. Complete the site register. Record every estimate, source gap, late return, or scope change in the exception log.
5. Complete the monthly attestation and submit the register, annexes, and exception log. Retain source records locally for audit; do not upload raw personal data unless separately requested through a secure, approved process.
6. Correct errors through a dated resubmission. Never silently overwrite a previous month.

## Core eligibility standard

For a month to feed the benchmark’s primary injury rates, all of the following must be true:

- The site was active and the Coverage Profile was current.
- Worker-hour and incident records cover the **same defined population**.
- At least **95% of submitted worker-hours** are source-supported by a time-and-attendance, payroll/timesheet, or certified subcontractor record.
- Estimated hours do not exceed **5%** of submitted worker-hours.
- All active in-scope workforce groups have a row or an approved, documented exception.
- The certifier has reconciled the month’s incident cases, including a verified zero-incident return where no case occurred.

A month that does not meet the standard is retained in the governed dataset and disclosed in data-quality reporting, but it is excluded from the primary rate calculation until corrected. The benchmark administrator may use a more conservative rule after methodology-panel review.

## Hours to include and exclude

| Include | Exclude |
|---|---|
| Actual hours worked by the covered workforce at the physical project site, including overtime. | Annual leave, sickness absence, rest days, and other non-worked paid time. |
| Site-based induction, toolbox talks, safety meetings, emergency drills, and authorised training where the covered person is at or assigned to the project site. | Commute, travel to and from the site, and off-site time unless the Benchmark Coverage Profile explicitly includes it and incident capture is equally in scope. |
| Project-site office, supervisory, and management time **only when those workers are included in the incident population**. | Remote office/design hours when those staff are outside the covered incident population. |
| Direct labour, labour-supply, and subcontractor hours where included in the Coverage Profile. | Hours belonging to a separate project, even if the organisation is the same. |

The International Labour Organization stresses that the reference group covered by injury data must be clear, and that relative measures are generally more comparable than gross injury counts.[1] The benchmark therefore requires worker-hours and incidents to share the same population and period.

## Source hierarchy

| Rank | Preferred source | Acceptable use | Not acceptable as the primary source |
|---:|---|---|---|
| 1 | Time-and-attendance, payroll, or approved digital timesheet export | Normal monthly submission | — |
| 2 | Certified subcontractor monthly hour return based on its payroll, timesheets, or time-and-attendance source | Subcontractor groups; attach annex and source reference | — |
| 3 | Controlled manual monthly summary with certifier declaration | Temporary fallback when system export is unavailable; include exception log | Routine permanent method without review |
| 4 | Estimate based on shift roster, headcount, or planned work | Only for a documented, limited exception; must be labelled `estimated` | Primary reporting method |
| 5 | Gate/access logs | Reconciliation and plausibility check only | Direct proxy for hours worked |

## Calculation and audit rule

The benchmark administrator calculates totals from the worker-hour register. Sites should not manually calculate rates in the submission pack.

```text
Total submitted worker-hours = sum(worker_hours) for all in-scope active workforce-group rows
Source-supported worker-hours = sum(worker_hours) where source_quality is actual_system, actual_payroll_timesheet, certified_subcontractor, or controlled_manual_summary
Estimated share = estimated worker-hours ÷ total submitted worker-hours
```

The benchmark publishes rates using worker-hours from eligible site-months only. For context, a common international convention expresses an incidence rate as injury cases × 200,000 ÷ employee hours worked. This convention represents the annual hours of 100 full-time workers; it is not a UAE legal metric.[2] The benchmark’s public primary rate is per **1,000,000 covered worker-hours**, as specified in its methodology.

## References

[1]: https://ilostat.ilo.org/methods/concepts-and-definitions/description-occupational-safety-and-health-statistics/ "International Labour Organization: Occupational Safety and Health Statistics"
[2]: https://www.osha.gov/laws-regs/standardinterpretations/2016-08-23 "Occupational Safety and Health Administration: Clarification on calculating incident rates"