MOP vs SOP vs EOP: What’s the Difference?

Overview

Choosing the wrong type is an operational problem, not a paperwork problem. A planned intrusive activity run under a standing SOP loses its prerequisite checks, its hold points and its back-out. An emergency handled by searching for a planned procedure loses time that the facility does not have.

This article compares the three, shows how they interact during a single event, and gives a decision route for selecting the right one. It does not teach you how to write or review a MOP — those are separate articles, linked below.

This article provides general operational guidance. Apply site-specific engineering review, risk controls, manufacturer requirements and applicable regulations before use.

The three document types at a glance

The distinction is easiest to hold if you read it as a question of predictability and authority. A MOP controls a specific planned activity with a known start, a known sequence and a defined return to service. An SOP controls an activity that recurs in essentially the same form under normal conditions. An EOP controls a response to an abnormal or emergency condition that was not chosen and cannot be scheduled.

Sites use different names — method statement, operating procedure, emergency response procedure, work instruction — and contracts sometimes impose their own naming. Compare the control content, not the title on the cover sheet. Where site or client terminology differs, map it explicitly in your procedure register rather than assuming equivalence.

MOP: planned, one-off, controlled

A MOP exists because a specific activity will change the state of live plant. It names the assets, the affected systems, the prerequisites that must hold on the day, the numbered sequence, the hold points, the testing and the back-out. Its scope is deliberately narrow: this activity, this plant, this window.

Two features separate a MOP from the other types. First, it accepts a temporary reduction in resilience — the document should state the reduced state, its expected duration and the conditions that end it. Second, it is approved for a particular execution, so the revision issued is the revision executed.

• Use a MOP when the activity is planned, intrusive or changes the state of a critical system.
• Use a MOP when redundancy, capacity or protection will be reduced for a period.
• Use a MOP when the sequence, tests and recovery need to be reviewed and authorised before the work starts.

SOP: repeatable normal operation

An SOP governs work that recurs in the same form: routine rounds, environmental readings, access control handling, alarm acknowledgement and triage, shift handover, routine BMS operations within defined limits. It is written once and used many times by competent people operating inside their standing authority.

The practical test is whether the conditions are stable enough that the same instruction remains valid every time. Where an SOP starts to accumulate site-specific caveats — 'unless the second chiller is on maintenance', 'unless the generator is under test' — that activity has outgrown the SOP and should be controlled by a MOP for the affected occasions.

• Use an SOP when the task recurs unchanged under normal conditions.
• Use an SOP when no planned reduction in redundancy or protection is involved.
• Do not use an SOP to authorise a switching, isolation or configuration change that requires case-by-case review.

EOP: abnormal and emergency response

An EOP is written for conditions you do not choose: loss of utility supply, UPS on battery, generator failure to start, loss of cooling in a data hall, water ingress, fire alarm activation, loss of BMS or EPMS visibility. It is opened when something has already gone wrong, so it is written differently from a planned document — immediate actions first, then stabilisation, then escalation, then recovery and reporting.

Two design rules follow from that. The first actions must be executable from memory or from a single page, because nobody reads a twenty-page document during a first-minute response. And the authority model must be explicit in advance: who acts without waiting, who is notified, and at what point operations, engineering, the client or emergency services are engaged.

Recovery after an event is normally not part of the EOP. Once the facility is stable, planned restoration work — returning plant to normal configuration, replacing failed components, re-proving redundancy — is usually controlled by a MOP prepared for that purpose.

• Use an EOP when the condition is unplanned and response time matters.
• Keep the immediate-action set short, unambiguous and available at the point of use.
• Define escalation by role and threshold, not by individual name.
• Return to planned control — normally a MOP — once the situation is stable.

How the three interact during a single event

The document types are not alternatives; a single sequence of events can move through all three. The following illustrative sequence is based on a common arrangement and is not a site-specific instruction.

A planned generator load test is controlled by a MOP: prerequisites confirmed, hall load and UPS autonomy verified, hold point acknowledged before transfer, defined stop conditions and a written back-out. During the test, the generator shuts down on an unexpected fault. The activity stops at the stop condition written into the MOP, and the EOP for loss of standby capability governs what happens next — stabilise supply, confirm UPS state, notify operations and the client, record the time. Routine monitoring, rounds and shift handover continue under their SOPs, because normal operation of the rest of the facility has not changed. Once the fault is understood, a new MOP is prepared for the repair and the re-test; the original MOP is not reopened and adapted in the field.

Choosing the right document type

Work through the questions in order and stop at the first that applies. The route below is a decision aid; the site's governance process remains the authority on which activities require which document.

• Is the condition unplanned and does it require immediate response? Use the EOP.
• Will the activity change the state of a critical system, remove redundancy or protection, or require isolation or switching? Use a MOP.
• Is this the same task, performed the same way, under normal conditions, by someone with standing authority? Use the SOP.
• Is it routine in principle but the plant is currently degraded, under test, or subject to concurrent works? Control it under a MOP for that occasion.
• Are you unsure? Escalate the selection decision before work starts — the point of doubt is the cheapest point to resolve it.

Who owns and approves each

Ownership arrangements vary by organisation and contract, so treat the following as a common pattern rather than a rule. MOPs are typically drafted by the party performing the work — in-house engineering or a contractor — then reviewed technically and operationally and approved by a named site role before issue. SOPs are usually owned by the operations function, because they describe how the site runs. EOPs are normally owned jointly by operations and engineering, with input from safety and, where relevant, the client, because they commit the site to a response.

Whatever the arrangement, three things should be unambiguous in the register: who may approve each type, what the review interval is, and which revision is current. Procedure sets fail more often through document control than through technical content — superseded revisions in circulation, procedures that reference decommissioned assets, and EOPs that have never been walked through after a change to the plant they cover.

• Record the approver by role for each document type, not by individual.
• Set a review interval for SOPs and EOPs, and review EOPs after every activation or drill.
• Withdraw superseded revisions from the point of use, not only from the document system.
• Re-check procedures against the asset register after any change to the plant they cover.

Frequently asked questions

Is a MOP the same as a method statement?

Not necessarily. A contractor method statement describes how that contractor intends to perform the work. A site MOP controls the activity within the facility's operational controls — release, plant state, prerequisites, hold points, stop conditions and return to service. A method statement is often an input to a MOP rather than a replacement for it.

Can one document be both a MOP and an SOP?

It is better to keep them separate. If a task is stable enough to run under standing authority, control it with an SOP. If a particular occasion carries additional exposure — degraded plant, concurrent works, isolation — raise a MOP for that occasion and reference the SOP where the technical method is unchanged.

Who writes EOPs?

Practice varies. EOPs are commonly developed by operations and engineering together, with safety and client input where relevant, because they commit the site to an immediate response and to an escalation path. Approval authority should be defined in the site's procedure governance.

How often should procedures be reviewed?

Set the interval in site governance rather than assuming a universal figure. In addition to any periodic review, revise a procedure after a plant or configuration change, after an activation or drill, after an execution that exposed a defect in the document, and when referenced assets or roles change.