The facility does not stop supporting science on Friday
It is Saturday afternoon. A cooling-system alarm arrives. On Monday morning, a research team expects to begin a key experiment—or manufacturing expects to start a scheduled production run.
The immediate question is not simply whether someone received the notification. It is whether the company has people who can assess the problem, understand what depends on the affected system, obtain the right support, and establish whether the facility will be ready.
A building-management system can send an alarm. It cannot supply missing judgment, spare parts, vendor availability, or authority to act.
Facilities staffing should be designed around the work the business must protect, including the time available to recover when something fails. Monday readiness is the result of decisions made well before Saturday.
A small HVAC fault can become a major scientific delay
An apparently simple HVAC issue—a failed actuator, a faulty sensor, or an air-handling unit that stops operating correctly—can affect much more than whether a room feels comfortable. Depending on the space and activity, temperature, humidity, ventilation, or pressure relationships may be important to safe work, experimental consistency, or production readiness.
Consider an illustrative Saturday afternoon failure in a room scheduled for a key Monday experiment. The repair itself may be straightforward. But if the problem is not assessed until Monday, the scientific team may arrive with prepared materials, reserved instruments, and a narrow experimental window—only to find that the room is not ready.
For a production area, restoring airflow may likewise be only the beginning. The team may need to assess the period of abnormal conditions, evaluate affected materials or operations, and complete the checks and approvals required for the intended use.
The size of the repair does not determine the size of the disruption. A minor component can sit upstream of an expensive study, a coordinated production schedule, or work that cannot simply be moved to the next day.
This is why facilities coverage must connect technical response with scientific priorities. Recognizing the affected room, understanding Monday’s dependency, and acting on Saturday can preserve time for repair, stabilization, and verification. Clearing the alarm alone is not the finish line.
Staff for consequences—not square footage alone
Two similarly sized facilities can require very different support. One may house flexible office and research activity. Another may support sensitive samples, tightly scheduled studies, controlled environments, and production processes that cannot simply restart without assessment.
Begin with the critical work and map its dependencies: power, cooling, ventilation, gases, water, environmental monitoring, equipment, and the interfaces between them.
For each important dependency, ask:
- How quickly would a failure affect people, materials, experiments, or product?
- How would the team detect it, and who would assess its significance?
- What skills and site knowledge are needed to respond?
- How long could diagnosis, repair, stabilization, and verification reasonably take?
- What alternative exists if recovery cannot be completed in time?
Use those answers to define coverage and recovery expectations. Do not assume a generic staffing ratio or a vendor’s standard service package matches the site’s needs.
Capability and bandwidth are separate requirements
A capable facilities lead may understand the building exceptionally well and still lack the capacity to cover preventive maintenance, projects, documentation, vendor supervision, and every unexpected interruption.
The opposite problem also occurs: several available people may lack the technical depth or site familiarity to resolve a consequential failure.
Build both dimensions deliberately. Identify the capabilities the team must retain internally, the specialist work that can be contracted, and the people who can provide competent backup. Cross-train on critical systems and keep current drawings, operating information, asset histories, and escalation contacts accessible.
Include planned leave, illness, training, simultaneous incidents, and recovery time after an overnight response when assessing staffing capacity. A plan that works only when the most knowledgeable person is always available is a single point of failure—not a coverage model.
An on-call number is not a recovery plan
After-hours coverage needs a defined sequence from detection to a documented decision:
- Acknowledge: A named responder receives the alert, with escalation if they do not respond.
- Assess: A qualified person determines urgency, affected operations, and immediate protective actions.
- Mobilize: The team can access the site, approved technical support, parts, and authorized spending.
- Communicate: Scientific or manufacturing owners receive timely information about impacts and options.
- Verify: Appropriate personnel confirm system performance and the status of affected work before restart.
Distinguish a commitment to answer the phone from a commitment to attend the site, and distinguish both from a realistic restoration capability. A contractor who can arrive Monday may not protect a Monday morning start.
Check the practical details before they matter: weekend dispatch arrangements, escalation contacts, access credentials, landlord responsibilities, critical spares, and purchasing authority. Include a backup route if the preferred responder or supplier is unavailable.
Not every facility needs a fully staffed weekend shift. Some need a well-supported on-call rotation; others need continuous on-site capability. Choose the model from the hazards, criticality, and recovery window—not from tradition.
Protect Monday without making heroics the operating model
A strong team may occasionally make an extraordinary effort. Leadership should appreciate that effort without building routine operations around it.
Repeated weekend callouts can signal inadequate staffing, deferred maintenance, poor alarm management, or unreliable infrastructure. Ask why the same system keeps interrupting the same people.
Budget for coverage, backup, and appropriate compensation. Manage fatigue and safe working arrangements; do not expect someone to complete a full workweek, respond through the night, and then resume normal duties without considering fitness for work.
After-hours urgency does not authorize unqualified repairs, bypassed safeguards, or work that should not be performed alone. The response plan must support safe escalation and, when necessary, a controlled delay.
Protecting the schedule is important. Protecting people and the integrity of the work comes first.
“Running again” is not the same as “ready to use”
Restoring a pump or clearing an alarm does not establish that the affected experiment or production process can proceed.
The team may need to examine environmental trends, assess an excursion, check impacted materials, verify stable performance, or complete additional testing and documentation. Include that time in the recovery plan.
For drug manufacturing subject to the relevant requirements, 21 CFR 211.67 addresses equipment maintenance procedures, assigned responsibilities, schedules, and records. It does not prescribe a universal facilities headcount; the staffing model must support the work and controls the operation requires.
Define decision rights in advance. Facilities confirms the technical condition of the system; scientific or manufacturing owners assess operational consequences; Quality participates in impact assessment and return-to-use decisions where required by the applicable procedures.
Do not let schedule pressure turn a technical repair into an unsupported release decision. Sometimes the correct outcome is a documented postponement, with stakeholders informed early enough to adapt.
Use Friday to reduce Saturday uncertainty
Before a high-consequence weekend or Monday start, review known issues with the teams whose work depends on the facility.
Confirm critical system status, relevant maintenance, unresolved alarms, upcoming work, and the escalation roster. Identify anything that could reduce recovery options, such as a spare awaiting delivery or a specialist unavailable that weekend.
This should be a focused readiness check, not a new meeting for every routine activity. The purpose is to surface risks while there is still time to address them.
A computerized maintenance management system can help make asset history and open work visible. It cannot replace the capability and capacity needed to act on that information.
Test the coverage model before the next interruption
Run a practical tabletop exercise: a critical utility fails at 3 p.m. Saturday, the primary facilities lead is unavailable, and important work is scheduled for Monday.
Walk through detection, backup response, diagnosis, vendor access, parts, approvals, communication, and return-to-use. Use realistic availability and lead times rather than assuming every call is answered immediately.
Identify where the plan depends on one person’s memory, an informal favor, or an approval that cannot be obtained. Assign those gaps to owners and test the revised arrangements.
Track response and recovery performance, repeat failures, maintenance backlog by criticality, and the load on the people providing coverage. Use the findings to adjust staffing, contracts, and infrastructure—not simply to demand faster responses.
Operational takeaway: fund readiness, not just headcount
Choose the next critical experiment or production run and identify the facilities systems it relies on—including the HVAC conditions the work requires. Then ask who will protect that readiness if a seemingly minor problem starts on Saturday afternoon, and how much time must remain for verification after the repair.
If the answer is a single person, an untested vendor promise, or “we will figure it out,” there is work to do.
Build a facilities team with the technical capability, available bandwidth, backup coverage, and authority to manage the likely failures. Give it the information, equipment, and specialist support to recover safely—and a clear process for deciding when the work can resume.
The objective is not a guarantee that nothing will fail. It is a business that can respond deliberately, protect its people and science, and make an informed Monday decision without depending on weekend heroics.
See this in practice
For a project spanning infrastructure and equipment startup and shakedown, see a laboratory relocation from Steve Michel’s prior leadership experience.