The maintenance report says 640 open work orders. The staffing request says six mechanics and four electricians. Nothing between those two numbers explains why those trades, why that crew size, or what they can start on Monday.
That gap is where backlog plans fail. A work-order count mixes unlike jobs: a five-minute record correction, an urgent pump repair, a motor replacement waiting on delivery, and a shutdown task needing several crafts may each count as one. A useful staffing plan starts by scoring the work, separating priority from readiness, and converting only executable hours into craft demand.
A large backlog is not yet a labor forecast
Backlog age and volume matter, but neither tells a maintenance leader what to staff. First remove duplicate requests, canceled modifications, completed work that was never closed, and inspection findings that should be linked to an existing order. Then make each remaining order specific enough to estimate. At minimum, record the asset, failure or condition, task boundary, required craft, estimated hours, operating constraint, material status, and completion test.
The U.S. Department of Energy's Federal Energy Management Program publishes an operations and maintenance best-practices guide that covers O&M management, preventive maintenance, predictive maintenance, and performance measures. It is a broad federal facilities guide, not a universal staffing formula. Its useful lesson here is structural: maintenance decisions need an operating program and measures, not a raw queue alone.
Keep two questions separate:
- How important is this work?
- Can a qualified crew execute it in the planning period?
A high-consequence job with no approved isolation plan may deserve immediate planning attention, but it should not create productive crew hours for next week. Combining importance and readiness in one vague "priority" field hides that distinction.
Score consequence with the CRAFT method
Use the CRAFT score to order maintenance planning effort. Score each work order independently, using definitions the plant approves.
| Factor | Question | Points |
|---|---|---|
| Consequence | What happens if the condition continues through the planning window? No material effect = 0; limited local effect = 1; production, environmental, quality, or asset exposure = 2; plant-defined highest consequence = 3 | 0 to 3 |
| Recurrence | Has the same failure or defect returned after prior intervention? | 0 or 1 |
| Age against due date | Is the order past its approved due date or approaching a fixed outage window? | 0 or 1 |
| Functional dependency | Does the task unblock inspection, startup, another craft, or a sequence of work? | 0 or 1 |
| Time sensitivity | Will delay make access, damage, scope, or recovery materially harder? | 0 or 1 |
The maximum CRAFT score is 7. Use score bands only after plant leadership defines them. A practical starting policy is 6 to 7 for immediate planning review, 3 to 5 for the next scheduling cycle, and 0 to 2 for normal queue review. The score does not override emergency procedures, regulatory obligations, manufacturer requirements, or site safety rules. Those controls remain separate gates.
CRAFT is deliberately narrow. It ranks attention; it does not pretend to calculate risk from a few fields. Record the reason for each point so planners can challenge inconsistent scoring instead of debating a colored cell.
Apply the READY gate before counting hours
Next, test whether the work can enter the staffing horizon. Mark one of these five READY conditions for every order:
- R, released: scope, authorization, parts, access, work controls, required information, and acceptance method are confirmed for the scheduled window.
- E, expected: one minor constraint has a named owner and dated resolution before crew assignment.
- A, awaiting material or vendor: required parts, rental equipment, or specialist support are not confirmed on site.
- D, definition incomplete: troubleshooting, drawings, engineering disposition, estimate, or task boundary is missing.
- Y, yield to operating window: the work is defined but cannot start until an outage, sanitation window, production release, or other approved access point.
Only R hours belong in committed staffing demand. E hours may support a separate upside scenario, but do not hire against them without a clear conversion date. A, D, and Y hours stay visible in the constraint plan.
This gate also forces coordination. OSHA's communication and coordination guidance calls for host employers, contractors, and staffing agencies to exchange hazard information and coordinate before work begins and as conditions change. READY does not certify safety or authorize work. It shows whether the planning inputs defined by the site are present.
Convert released hours with the staffing worksheet
Build one row for each craft, shift, and planning period. Do not combine day-shift electrical work with night-shift mechanical work simply because both fall in the same week.
| Worksheet field | Example: day-shift electricians | How to calculate or decide |
|---|---|---|
| Released direct hours | 360 | Sum estimated hours for R-coded orders assigned to this craft and shift |
| Add approved emergent reserve | 36 | Plant policy; here, 10% of released hours |
| Total direct-hour demand | 396 | Released hours + reserve |
| Paid hours per person | 40 | Scheduled hours in the period |
| Direct-work factor | 0.70 | Share of paid time planned for direct task execution after meetings, travel within the site, breaks, coordination, and normal delay |
| Direct hours per person | 28 | Paid hours × direct-work factor |
| Base people required | 14.14 | Total direct-hour demand ÷ direct hours per person |
| Planned craft headcount | 15 | Round up, then test supervision and workfront limits |
The example is arithmetic, not a benchmark. A plant should derive its direct-work factor from its own time coding, schedule history, or a short measured study. Do not inflate the factor to make a requested headcount fit. Also keep the emergent reserve explicit. If routine break-in work consumes far more than the allowance, investigate the asset strategy and planning process rather than quietly borrowing capacity from scheduled work.
Repeat the worksheet for millwrights, instrumentation technicians, electricians, welders, planners, and any other required capability. Split specialist work from general craft hours when qualifications, tools, or authorization differ. Then add leads and supervisors according to the employer's structure, site rules, task complexity, shift coverage, and number of simultaneous workfronts. There is no responsible universal ratio.
Reconcile the math against physical workfronts
Fourteen calculated people cannot be productive if the plant can open only two small work areas. Before finalizing the request, map the proposed crew to actual workfronts. For each workfront, name:
- the orders and equipment boundary;
- the craft mix and lead;
- the shift and access window;
- the isolation, permit, and operating interfaces controlled by the site;
- staged material, special tools, lifting support, and inspection needs;
- the person who accepts the completed work and closes the record.
If calculated demand exceeds usable workfront capacity, choose deliberately. Extend the horizon, add a shift with full support, move work into a planned outage, or accelerate constraint removal. Adding people without another executable workfront usually adds coordination load before it adds completed maintenance.
Decide what the internal team must retain
Plant employees often hold asset history, troubleshooting judgment, operating authority, and relationships with production. Protect that capacity. A supplemental crew is easier to integrate when the package has a clear boundary and completion standard, while internal experts retain diagnosis, prioritization, site authorization, and acceptance where appropriate.
OSHA's temporary-worker guidance explains that staffing agencies and host employers share responsibility for maintaining a safe work environment, with the extent depending on the facts of each case. Before placement, define who supplies general and site-specific training, who communicates changed hazards, who supervises the task, and who workers contact with a concern. The site's controlling procedures and applicable requirements determine the details.
When the released-hours worksheet shows a real capacity gap, Rinvio's W-2 crew approach provides one option for skilled-trades staffing. Send the workfront, craft, shift, start window, site requirements, supervision plan, and expected duration. That packet is far more useful than sending the backlog count.
Review completions and constraints every week
After mobilization, compare planned direct hours with accepted completed hours by craft and workfront. Also track orders released, orders completed, estimated versus actual hours, hours lost to named constraints, emergent work absorbed, rework, and overdue closeout records. These measures show whether the plan is feeding executable work to the crew.
The weekly decision is simple: release more work, remove a constraint, rebalance a craft, change the workfront, or adjust capacity. Staffing follows that decision. The backlog count remains useful context, but the worksheet turns it into something a maintenance leader can defend, a planner can update, and a crew can execute.
Sources
- Operations and Maintenance Best Practices Guide: Achieving Operational Efficiency, U.S. Department of Energy Federal Energy Management Program, accessed July 12, 2026.
- Communication and Coordination for Host Employers, Contractors, and Staffing Agencies, Occupational Safety and Health Administration, accessed July 12, 2026.
- Protecting Temporary Workers, Occupational Safety and Health Administration, accessed July 12, 2026.
Featured image: EU-Ukraine cooperation, via Flickr (CC BY-SA 2.0).