Maintenance Shutdown Planning: Scope, Schedule, Critical Path

What a maintenance shutdown is

A maintenance shutdown (also called a turnaround or planned outage) is a planned interruption of production, with a fixed date and duration, to carry out work that cannot be done while the equipment runs: internal inspections, replacement of major components, overhauls and regulatory upgrades. In process plants it is often called a turnaround.

Because every hour of downtime costs margin that does not come back, the job splits into three: a scope frozen early, a schedule with a known critical path, and execution with change control.

Visual map of the maintenance shutdown

The map shows the phases, the 58-hour schedule with its critical path, the hourly cost and the close-out, using the same numbers as this article.

Scope: what goes in and what stays out

A shutdown that goes wrong usually has an open scope. Build the work list from four sources:

• Maintenance plan: preventive tasks that need the machine stopped, as in the maintenance plan.
• Corrective backlog: open work orders that can only be done with the equipment out of service.
• Inspections and statutory requirements: periodic inspections of pressure vessels, boilers and safety devices, as required by the rules that apply to the equipment.
• Improvements and modifications: only those that already have a design, materials and approval.

Each item enters with criteria: why it needs the shutdown, how long it takes, which materials and which trades. Set a scope freeze date (for example, 60 days before) and a formal process for exceptions. Organize the list in a work breakdown structure by system or equipment.

Schedule and critical path: a 58-hour example

Example (fictional): shutdown of a process line. Durations in hours; ES and EF are earliest start and finish; float is the difference between latest and earliest start.

Critical path: A → C → D → F → H → I = 6 + 10 + 24 + 8 + 6 + 4 = 58 hours. A delay in any of these delays the whole shutdown, while the electrical work (G) has 18 hours of float. The method is explained in PERT/CPM, and the schedule can be shown as a Gantt chart.

Cost of the shutdown and a window with reserve

Assume a contribution margin of $12,000 per hour of line downtime (see contribution margin). The 58-hour shutdown costs 58 × 12,000 = $696,000 in lost production, not counting services and materials.

Cost of delay = hours of delay on the critical path × margin per hour

If the production window is 66 hours, the reserve is 8 hours (about 14% over the 58). A 6-hour delay in D ends the shutdown at 64 hours and still fits the window; the 8 reserve hours protect $96,000 of production. A 6-hour delay in E (finish at hour 42) uses up the 4 hours of float and pushes the shutdown to 60 hours: the cost is 2 × 12,000 = $24,000.

Resources and schedule compression

To shorten the job, act only on the critical path. A second crew on D cuts the activity from 24 to 18 hours, and D finishes at hour 34. But E finishes at hour 36, and the critical path becomes A → C → E → F → H → I: 6 + 10 + 20 + 8 + 6 + 4 = 54 hours. The shutdown drops by 4 hours ($48,000), not 6, because another path took over.

Before compressing, compare the cost of the extra crew with the gain. Also check crane and laydown-area availability, the number of people per work front, materials verified and on site before the shutdown, and contractors who are qualified and trained in the plant rules.

Execution and control

1. Pre-shutdown: kick-off, materials on site, pre-assembly, safety induction and the isolation plan. Apply lockout/tagout by equipment and by crew.
2. Shift handover meeting: compare actuals with the schedule, review critical activities and the next shift's risks.
3. Progress curve: track percent complete against plan, also through earned value and the S-curve.
4. Change control: every new job (an inspection finding) gets an impact assessment on the critical path before it is accepted.
5. Risk management: list risks such as late deliveries of parts or weather, with a response plan (risk matrix and preliminary risk analysis).

Close-out and lessons learned

The shutdown ends when the equipment starts up stable and the paperwork is closed: work orders recorded in the CMMS, inspection reports filed, deviations and deferred jobs logged. Within a few weeks, hold the lessons-learned meeting and measure: schedule adherence (actual hours ÷ 58), scope adherence (jobs done ÷ planned), actual cost against budget, and failures in the first month after start-up, which you can track with MTBF and MTTR.

Common mistakes

• Scope that grows unchecked, with jobs added in the last days.
• Late materials: the critical path stops while waiting for a part.
• A schedule without a critical path: every activity is treated as urgent and none is prioritized.
• No reserve: the date is set from the minimum duration and any surprise blows the window.
• Too many people on one front: space and safety limit the gain.
• Skipping supervised start-up and no-load tests to save time.
• Not recording lessons and repeating the same delays in the next shutdown.

Frequently asked questions

How far ahead should I start planning a shutdown?

It depends on size. Large shutdowns need months to close scope, buy long-lead materials and contract crews; short ones can be planned in weeks.

What is the critical path in a shutdown?

The sequence of activities with zero float that sets the total duration. In the example: A, C, D, F, H and I, at 58 hours.

How do I calculate the cost of the shutdown?

Hours of downtime × contribution margin per hour, plus the cost of services and materials. In the example, 58 × $12,000 = $696,000 of lost production.

Why freeze the scope?

Because each new job affects materials, people and schedule. The freeze forces an impact assessment before an addition is accepted.

Is it worth shortening the shutdown with more people?

Only on the critical path and only if the gain exceeds the extra cost. In the example, a second crew on D saves 4 hours ($48,000), not 6, because the E path becomes critical.

What about jobs that did not fit?

Log them as backlog in the CMMS with risk and deadline, and include them in the next shutdown or a short outage window.

About the author

Vagner Soares

Lean Manufacturing & Behavioral Management Specialist

Over 20 years in the automotive and metalworking industries (GM and Dana), Lean Manufacturing practitioner since 2006. SENAI instructor and mentor in Brazil’s Brasil Mais Produtivo program, delivering consulting, training and audits for 50+ companies, combining quality, productivity and people development.

LinkedInInstagramFull profile →