What we learned running the KiwiRail EBS pilot

On 16 September 2026 we ran an Energy-Based Safety pilot workshop for KiwiRail Wellington Metro, delivered with CHASNZ.

It was a pilot. We are not going to tell you it changed KiwiRail’s injury rate, because one day in a room does not do that and we have no data that says it did. What we can tell you is what the day looked like and what we took from it, which is useful if you are weighing up a pilot of your own.

Why a pilot, and why this group

KiwiRail had been to a CHASNZ introduction to Energy-Based Safety two years earlier. The original ask was a repeat. CHASNZ suggested the blended format instead, with the VR scenarios, and positioned it as a pilot.

The group mattered. H&S reps and the risk team are the people who will carry a programme into the business or quietly let it die. Running the pilot with them first, rather than with a frontline crew, meant the people in the room could judge whether the method fits rail and whether they could teach it themselves.

How the day ran

1. CHASNZ opened with the general introduction to Energy-Based Safety. Because that overlaps with the first workshop, we blended the two rather than making people sit through the Energy Wheel twice.
2. Workshop one: Energy-Based Hazard RecognitionThe headsets came out here. Participants scanned a scene by instinct, then again with the Energy Wheel, and compared what they found. The second pass finds more, and watching it happen to yourself is more convincing than being told.
3. Workshop two: High-Energy Hazards and Direct ControlsThe 1,500 joule threshold, the high-energy examples, and the three-part test for a Direct Control. Two VR modules follow. In the first, people judge three high-energy situations: how likely, then how confident. In the second, they test the controls on a site against the 3Ts, then Strong and Forgiving. The arguments start here.
4. Workshop three: HECA, the High Energy Control AssessmentOne task, every high-energy hazard listed, each one scored as a success if it has a Direct Control or an exposure if it does not. Pairs scored the same worked example and compared their scores, which exposes how differently two sensible people can read the same control.

The rail questions were the useful part

The programme grew up in construction and utilities, and the scenarios reflect that. The best moments of the day were when the room translated it into their own work.

• Track work with trains running
• Working under a raised wagon
• Overhead line work
• Shunting
• Platform edges

Each of these had to be placed on the Energy Wheel, measured against the threshold and tested for a Direct Control. Some had obvious answers. Some did not, and the discussion about what a Direct Control even looks like for a particular rail task was exactly the discussion the programme exists to provoke.

That is the point we would make to anyone in a sector the framework was not written for. The scenarios are a starting point. The transfer happens when your people argue about your tasks.

What we took away

Timing

Each workshop is designed for about 90 minutes and runs longer when there is more VR in it. Compressing three into a day is possible with a group this size, but the first workshop has to be blended with the introduction and the others need their full time.

Group size

Around ten people with a facilitator and a partner on hand is a comfortable number for headsets. Everyone gets scenario time and the discussion stays whole-room.

The reps and risk team as the first audience

This worked. The people who will have to defend the programme internally got to test it on their own hazards first.

Position it as a pilot

Saying so up front took the pressure off the day to be a finished rollout and made the conversation about what should come next honest.

What comes next

We hope to be back in the classroom with KiwiRail, this time with VR experiences tailored to rail.

If you want to see the method with your own crews, scope a pilot.