Serious injury and fatality prevention that finds the missing critical control in today's high-energy work, before the crew leaves the yard
Every high-energy job checked for its critical controls before the crew leaves the yard.
See one case, screen by screen ↓Recordable injury rates in power generation and delivery fell from about 2.05 to about 1.14 per 200,000 hours between 2010 and 2023, while the fatality rate stayed flat at about 0.02 per 1,000 workers (EEI, High-Energy Control Assessments, 2024). Serious injuries and fatalities do not come from the same causes as sprains and cuts. They come from high energy — an energised 12.47 kV line, a climb to 38 ft, a 7 ft trench in wet clay, a chlorine cylinder, a vault entry — meeting no control that works when someone makes a mistake.
The controls are known: fall protection on a climb, a shield rated for the depth and soil, breathing apparatus at a chlorine changeover, an attendant outside a permit space, verified isolation. What fails is that one control is not on one truck on one morning — a fall restriction device tagged out last Friday, a pack built from a bucket template for a pole no bucket can reach, “shoring as required” with no shield assigned. With 214 jobs and 11 supervisors, no one can check that before 07:00. Supervisors read tailboards on site after the crew has arrived, and visits are chosen by habit and distance, not by exposure.
Estimated split, in calendar days, from a near miss to packs built from a corrected template — by hand and with the solution. In the working solution the rear-lot climbing pattern on template OH-07 runs from a near miss on 21 Jul to today's red job on 13 Oct.
Six specialist agents find each job's high energy and check its controls before work; the field supervisor decides.
Tuesday 13 October, Lumina Utilities, Corran Valley. 214 jobs — 152 electric, 62 water — leave four yards at 07:00. Here is what happens before they do, screen by screen, in the working solution.
The agents read 214 work orders and their packs, 38 truck manifests, 162 crew records and the 05:30 forecast. 61 jobs meet high energy, with 148 hazards; 12 are worked out in joules from pole heights, loads and depths. 136 have a working control, 5 are listed but not verified, and 7 have none, on 5 jobs. The map shows each high-energy job by its main energy, coloured by its controls.
Precursor scanner: “Gusts 31–33 mph from 12:00 on the north ridge · 3 scopes changed since the packs were built · 2 crews worked over 11 h yesterday”
Hannah opens the map nineteen minutes before the crews leave. Five jobs are missing a control. She opens WO-48213: replace a crossarm and insulators on pole P-11873, feeder KM-04, 12.47 kV energised, Crew 14, start 07:30. Electrical contact, arc flash and the dropped crossarm each have a control in place. Fall from height — a climb to 38 ft, about 10,200 J — does not.
“Fall-arrest equipment is not listed in the pack, and truck T-4471's pole fall restriction device was tagged out on 9 Oct.”
The job screen lays every high-energy hazard beside the control that works even if someone slips, the pack line and the record that verifies it: gloves and sleeves class 2 tested 21 Sep, 8 cal/cm² clothing against about 4.1 cal/cm² at working distance, a barricaded drop zone. The fall line reads “Not in the pack”. The supervisor decision opens on the recommended fix: move two fall restriction devices from T-4490.
A job missing a control may not start until Wes Calder amends the pack, holds it, or signs a go with a reason.
The pack was built from template OH-07, crossarm replacement, energised (bucket), version 8. Its equipment list has “Bucket fall arrest, anchored to the boom”. But P-11873 is in a rear lot 64 ft from the road on a 14 % slope, so the crew climbs. Insulators were added at 16:20 yesterday and the equipment list was not rebuilt. Jamie Reyes, a second-year apprentice, is on the climb — a first rear-lot climb on record, and gusts pass the 30 mph climbing limit at 12:00.
Margin note: “No fall-arrest equipment for climbing — and this pole has no bucket access”
The alert reaches Wes's phone at 06:42 with the job, the missing control and its source. He acknowledges at 06:43. At 06:58 he moves pole fall restriction devices FRD-118 and FRD-121, inspected 30 Sep, from T-4490 — Crew 12 is on ground work today — to T-4471, adds them to the pack and re-sequences the climb to finish by 11:30, before the gusts. The checker checks again: 4 of 4. Start moves to 07:50.
Wes Calder: “On it. Taking the devices off 4490.”
Back on the map, the Kilnmoor pin is green and the count reads 137 of 148 hazards with a working control. The job's log keeps the alert, the acknowledgement and the amendment with their times. Four jobs are still red: the water main break in Pellham Cross, the chlorine changeover at Ashcombe, the 11 kV vault in Wrenfield and the 50 ft pole beside a live 34.5 kV line in Tarrow.
“WO-48213 is green · 137 of 148 hazards now have a working control”
WO-48177 repairs a 12-inch ductile iron main outside 41 High Street, Pellham Cross. The trench is 7 ft deep in type C clay, wet after 0.7 in of rain, with excavator EX-14 working beside people on foot. The pack says “shoring as required” and no shield is assigned. Shield TS-3 is free at the Pellham Cross depot; the recommended amendment puts it on the low-loader for 08:20.
Never counts on its own: “shoring as required”, a spotter at the edge.
40 visits for 11 supervisors, with the 9 red and amber jobs first, each before its start. Wes has 4 visits and 77 minutes of driving from Kilnmoor yard: P-11873 at 07:40 to confirm the fall devices on site, then the Tarrow pole set at 09:50, where no one is yet named to watch the boom's distance to the 34.5 kV line. Green high-energy jobs fill the rest of the day.
Visit planner: never removes a visit a supervisor chose.
Wes arrives at P-11873 and checks each hazard as he sees it, before anyone steps in: gloves and sleeves on with cover-up on all three phases, 8 cal/cm² clothing, the drop zone taped, and both climbers wearing FRD-118 and FRD-121 from the ground. The observed control score is 100 % — 4 of 4 hazards — and matches what the morning check now plans.
Observed control score = controls in place ÷ hazards seen
The event classifier sorted 47 incidents and near misses from the last 90 days by high energy, control and outcome. Three on OH-07 share a pattern: an apprentice slipped at 12 ft with no device on 21 Jul, a climb was stopped on a visit on 24 Sep, a gaff cut out at 22 ft on 2 Oct and the device held. Version 9 adds “if the pole has no bucket access, the job is a climb” and a fall restriction device for each climber. It waits for Hannah's approval.
“This month 14 jobs use OH-07 on poles with no bucket access. Packs built after approval include the device.”
The same morning, seen by the five people who own it — what their day looked like, and what it looks like now.
Six specialist agents find each job's high energy and check its controls before work; the field supervisor decides.
At 06:00 reads every work order scheduled for the day with its job pack — scope, template and version, equipment list, permits, switching orders and hold-offs — and each assigned truck's equipment manifest and tag-out records. Records what the pack says and what is actually on the truck.
Works out the energies each job will meet from the work type, the asset and the site: circuit voltage and state, height of work and access, loads, trench depth and soil, chemicals, confined spaces, traffic speed. Where no threshold applies it computes the energy — a climb to 38 ft is about 10,200 J against the 1,500 J line.
Matches every high-energy hazard to a control in the pack that is targeted at the energy, works when installed and used properly, and works despite a human error. Checks the item is on the assigned truck, not tagged out, within its test or inspection date, and that the person using it is qualified.
Checks each high-energy job against the hourly forecast and weather limits, hours worked in the last 24, apprentices on high-energy tasks, scope changes after the pack was built, site access, and incidents or near misses on the same template or asset in 90 days.
Builds each supervisor's route from their yard: red and amber jobs first, each before its start time, then green high-energy jobs by drive time. Plans again when a job changes colour.
Reads each incident and near-miss report and decides whether high energy was present, a working control was in place, energy was released and anyone was seriously hurt. Classifies it as serious injury, serious potential, control held, exposure found, success or low energy, links it to the job template and proposes a change when a pattern repeats.
Decides every job missing a control; safety manager approves rule changes. The agents propose; a named person decides.
Ask about any job, energy, control or supervisor — or say what to hold. Every answer points to its records.
Why is pole P-11873 in Kilnmoor red?
Crew 14 has to climb: the pole is in a rear lot 64 ft from the road on a 14 % slope, so no bucket can reach it. The pack was built from template OH-07, which assumes a bucket, and lists no fall-arrest equipment for climbing. Truck T-4471's pole fall restriction device was tagged out on 9 Oct for a worn strap. The climb is to 38 ft — about 10,200 J — and gusts pass the 30 mph climbing limit at 12:00. Fastest fix: move FRD-118 and FRD-121 from T-4490, which is on ground work today.
Which trenches today are deeper than 5 feet?
One of today's trench jobs has no protection: WO-48177 on High Street, Pellham Cross, is 7 ft in wet type C clay and no shield is assigned. Shield TS-3 is free at the depot. The map is filtered to trenches.
Plan Wes Calder's visits for this morning
Wes has 4 visits and 77 minutes of driving from Kilnmoor yard. Both red jobs come first, each before its start: P-11873 for the fall devices before the climb, then the Tarrow pole set for an observer for the boom near 34.5 kV. His plan is open; send it to his phone when you are happy.
Hold the Pellham Cross main repair until a trench shield is on site
Held. WO-48177 (Crew W-6, 12-inch main on High Street) cannot start until trench shield TS-3 is on site and set. Marta Kowalski and Aisha Bello are told, and the valve operations booked for 07:40 move to 08:20. The hold is on the audit trail; releasing it needs Aisha or you.
13 screens from the working solution, on its sample data. Pick one to see it large.
What counts as high energy, what counts as a control, when the check runs and who it calls are settings — every change versioned, approved by the safety manager and on the audit trail.
| Setting | Default | Choose from |
|---|---|---|
| Energy that can kill | 1,500 joules | Above this, the most likely result of contact is a serious injury or a death; also 50 volts for electrical contact and 30 mph for traffic |
| Fall from height | 4 ft | Adjustable up or down — climbing, buckets, roofs, tanks, open edges |
| Excavation or trench | 5 ft | Adjustable up or down — the depth at which a protective system is required |
| Climbing poles and towers | 30 mph gusts | Adjustable — above it a climb is a precursor and is re-sequenced or held; bucket and boom work 30 · 35 · 40 mph gusts |
| Morning check | 06:00 every working day | 05:30 every working day · 06:00 every working day · 05:00 and 12:00 — and again when a pack, crew, truck or forecast changes |
| A red job may not start | On | On · Off — until its supervisor amends the pack, holds it, or signs a go with a reason |
| Who is alerted | Operations manager after 15 minutes | The crew's supervisor at once, then the operations manager at once · after 10 minutes · after 15 minutes |
| Treat as missing | On | On · Off — a control whose test, inspection or calibration is out of date |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample data · “target” = the design goal, measured in the pilot · “estimated” = our estimate · “typical” = published figures (DEKRA study of 1,060 recordables at 11 utilities, reported by Safety+Health, 2018 — 32 % with SIF potential, 42 % at water utilities; BLS Monthly Labor Review, 2018 — power-line installers and repairers, 19–24 fatal injuries per 100,000 full-time workers, 2011–2015). Context: EEI, High-Energy Control Assessments (2024) and Power to Prevent SIF — 1,500 J high-energy line, direct controls, recordables down while the fatality rate stayed flat; 29 CFR 1910.269(g), 1926.652, 1910.146. Lumina Utilities, the Corran Valley, and the people, crews, trucks and jobs named on this page are fictional — characters and sample data in the working solution.
Software that looks for the conditions that lead to serious injuries and deaths — high energy without a working control — instead of counting injuries after the fact. Serious Injury and Fatality Prevention checks every job of the day for its high-energy hazards and their controls before the crews leave the yard, and puts the gaps in front of the supervisor who can fix them.
At 06:00 six agents read every work order, job pack, truck manifest, crew record and the forecast. They work out the energies each job will meet, check each one has a direct control that is listed, on the truck, in test and used by a qualified person, and flag precursors such as gusts over the climbing limit or a changed scope. Jobs missing a control turn red on the map and the supervisor is alerted.
A control targeted at the energy, that works when installed and used properly, and works even if someone slips — a pole fall restriction device for a climb, a shield rated for the depth and soil, breathing apparatus for each person at a chlorine changeover. Training, signs, “as required”, cones alone or a spotter who also works never count on their own. The control library is a setting.
Yes. In the working solution the same morning check covers water main repairs and trenching, chlorine cylinder changeovers, lift-station pump work and valve work in live lanes, alongside overhead, substation and underground electrical work, including vault entries. The thresholds — 1,500 J, 50 V, 4 ft, 5 ft, 30 mph — apply to every job.
No. The precursor scanner flags conditions — weather, hours worked, an apprentice on a high-energy task, a changed scope, access — never people. Crew data is limited to qualifications and hours, and the event classifier never assigns blame.
Yes. The agents check and alert; they never edit a pack, start or stop work, or close an investigation. A red job may not start until its supervisor amends the pack, holds it, or signs a go with a reason, and stop-work authority stays with people. Rule and template changes need the safety manager's approval.
Every morning it measures planned controls — high-energy hazards with a working control — for every job, and every on-site check records the observed control score, controls in place divided by hazards seen. The trend shows planned controls daily and observed controls weekly, by yard, next to serious events by quarter. Serious events are too rare to read in a quarter or two; the daily measures move first.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your safety rulebook, job templates, control library and a sample of work orders, packs and near-miss reports — and it goes live once every quality gate has passed. We will walk you through it on a day of your own jobs first.
High-potential incidents found the hour they are reported, not at the monthly review.
1-min demoElectronic Permit to WorkNo permit goes live over a held or missing isolation, and every signature is on record.
1-min demoNear Miss ReportingRising near-miss trends caught by site and work type, weeks before someone is hurt.
1-min demoWe'll run Serious Injury and Fatality Prevention on a day of your own work orders and job packs.