HSE Lab
Concept preview · learning content in development

Check the data.
Understand the risk.
Explain the next action.

A professional learning space for HSE practitioners. Explore a dummy dashboard, test the evidence, and practise turning routine HSE records into a clear, well-reasoned recommendation. No programming is required.

Open the decision tool View the Lab 01 preview Open the demo dashboard
The method

A dashboard is the starting point. Your reasoning is the skill.

Many HSE reports show numbers. Fewer help people decide what to do next. Every lab follows the same three moves.

01

Check the data

What does the number actually mean? Look at definitions, missing records, exclusions and the evidence behind a result.

02

Understand the risk

What deserves attention first? Find patterns, link them to controls, and note what the records cannot tell you.

03

Explain the next action

What should happen next? Write a short management brief with a finding, a limitation, an action and a decision.

Lab 01 preview

Actions closed. Risk controlled?

You are preparing for a management review. Most corrective actions are marked closed, but similar hazardous events keep appearing. Management asks: “Are our actions working?”

Illustrative dummy data

Corrective action snapshot

46 / 50actions marked closed
12 / 46closed with effectiveness verified
4similar events in 6 months
Closed
92%
Effectiveness verified
26%

Percentages: 46 of 50 closed; 12 of 46 closed actions verified. Fictional numbers made up for this exercise. They are not from any real site or company.

Your move

Which statement is best supported by this data?

Write the management brief

Example structure. In a real lab you write your own, then compare.

Finding
46 of 50 actions are closed. Effectiveness is verified for 12 of the 46.
Limitation
We cannot yet say whether the four similar events share the same hazard and failed control.
Next action
Review the repeated events against the original hazard, correction type and verification evidence.
Decision required
Agree an owner and clear criteria for verifying effectiveness.
Interactive decision tool

From your numbers to a priority

Enter your lagging results (what already happened) and leading indicators (what you do to prevent it). The tool checks your data, shows where the biggest gaps are, and suggests what to investigate or invest in first. Then it drafts a management brief.

Your numbers stay on your device. This page does not send, save or store anything. Closing or refreshing the page clears it. Please use dummy, rounded or fully authorised figures. Do not enter confidential or personal information.
Scope. These indicators suit everyday injury and action management. They are not barrier-based process-safety indicators for major-hazard sites (see HSE’s HSG254 for that).
1. Context
Actual hours worked (exclude leave), for the same people and period as the cases you count.

Definitions of recordable, lost-time and closed differ between organisations and regulators. Use yours, and keep them the same from period to period.

2. Lagging results (what happened)
Include lost-time cases. Use your own recordable definition.
A subset of recordable.
Same hazard or failed control.
Target for recordable rate.
3. Leading indicators (prevention)
Hazards and near misses.
Of those reported.
Raised in this period.
Marked complete.
Checked that it worked.
Still open, due date passed.
Optional. Helps test whether stronger controls were considered.
Adjust thresholds (illustrative, set your own)

These are not industry standards. They are starting points for discussion. Change them to match your own targets.

Method and sources

What the tool uses, and where each idea comes from. Thresholds, scores and the ranking are this tool’s own rules of thumb, not published standards.

  • Injury rate: cases × base ÷ hours worked. The 200,000 base is 100 workers × 40 hours × 50 weeks (OSHA interpretation letter, 2016). Use actual hours worked, excluding leave (BLS). Choose one base and keep it consistent.
  • Recordable and lost-time: days away from work is one of the recording criteria, so lost-time cases are part of recordable cases (29 CFR 1904.7). Your organisation’s definitions may differ.
  • Closed is not the same as effective: ISO 45001:2018 clause 10.2 asks organisations to review the effectiveness of corrective action.
  • Control strength: NIOSH ranks elimination, substitution and engineering controls above administrative controls and PPE because they depend less on human action, while noting that administrative controls and PPE can still be effective (NIOSH hierarchy of controls).
  • Leading and lagging indicators: relying on injury data alone is discouraged (HSE HSG254). Hopkins (2009), Safety Science 47, 460–465, examines the lead/lag distinction critically. Reiman and Pietikäinen (2012), Safety Science 50, 1993–2000, separate outcome, monitor and drive indicators. The tool therefore asks whether leading indicators test control performance, not only activity.
  • Small numbers: 95% intervals use the exact Poisson method for counts up to 500 and Byar’s approximation above that (Public Health technical guidance: Rates). One injury moves a small site’s rate far more than a large site’s (OSHA letter above).
How people learn here

From idea, to practice, to evidence you can keep

HSE Lab is designed alongside a companion book on digital HSE skills. The book explains the ideas. The lab is where people practise them.

1

Read

Get the core idea from a book chapter.

2

Practise

Explore a dummy dashboard in a guided lab.

3

Worksheet

Complete one worksheet for the session.

4

Brief

Write a short management brief you can reuse.

Learning logic: Knowledge → Competence → Evidence → Visibility. Every session ends with something a participant can actually show.
Planned sessions

One lab and one worksheet for each session

Each session can be used on its own for a team sharing, or followed as a series. Titles and order may change as the content develops.

PLANNED

Session 0

Where do we start?Digital HSE readiness check.OutputReadiness gaps and a 90-day sprint
PLANNED

Session 1

Can we trust the numbers?Data quality checklist.OutputA data quality score and your top gaps
PREVIEW

Session 2

Actions closed. Risk controlled?Dashboard and predictive safety canvas, with the decision tool above.OutputRisk signals and the decision needed
PLANNED

Session 3

Does the dashboard drive decisions?Dashboard design assessment.OutputAn effectiveness score and a layout plan
PLANNED

Session 4

Where could AI or tech help?Digital opportunity planner.OutputA pilot choice and a 90-day plan
Intended audience

For practitioners who turn HSE records into decisions

Who it supports

  • Early-career HSE staff: evidence of practical skills.
  • Practitioners: a method for checking data before presenting it.
  • Supervisors and managers: practice in explaining a decision.
  • Facilitators: a structured lab and worksheet for a sharing session.

Approach

  • Starts with a question, not a software menu.
  • Separates “action completed” from “effectiveness verified”.
  • Uses imperfect, clearly labelled dummy data.
  • Ends with a brief, not just another chart.
Stay informed

Register your interest and share your views

HSE Lab is at concept stage. Register your email to receive updates and invitations to sessions, and tell us which topics would be most useful in your work.

Topics where your views would help:
  1. Which session would you attend first?
  2. Was the decision tool useful? What number or question was missing?
  3. Would you prefer a live workshop, a recorded lesson, or a printable worksheet?
  4. What HSE data question do you struggle with most?
Registration form coming soon

Your email address is used only to send updates about HSE Lab and invitations to sessions. Please do not enter confidential or company information in the form.

About

About the author

I’m Azhara, an HSE practitioner interested in making everyday data more useful for decisions. HSE Lab brings together dashboards, structured thinking and clear management communication.

Practice safely. Use synthetic or explicitly authorized data only. Please do not share confidential records, personal information or internal incident details.