Planning maintenance with a Gantt chart

A Gantt chart is a way of seeing work over time: each task is a bar, whose position says when it happens and whose length says how long it lasts. It was born to manage projects, but it solves a different, very everyday maintenance problem quite well: distributing people across jobs.

The reason is that a list doesn’t let you see two things a chart shows at a glance: overlap and gaps.

What problem it actually solves

The pattern that repeats across almost every department is the same. Half the team overloaded and the other half waiting, and nobody notices until someone complains or until a job runs late.

With a list of orders sorted by date, that’s invisible, because the list has no axis of people. With bars laid out by technician, you immediately see who has a full week, who has three gaps, and where a long job is going to collide with a scheduled preventive visit.

What has no owner

The most useful information on a planning screen isn’t what’s assigned: it’s what isn’t. A job with no technician belongs to no one, and what belongs to no one falls through silently.

In GMAO Cloud’s calendar you can tell unassigned work apart by color for exactly that reason. It’s a visual detail with real consequences: what’s visible gets distributed, and technicians themselves can split work better among each other when they’re looking at the same picture.

Rescheduling by dragging

A maintenance plan changes every day. An urgent breakdown comes in, a client won’t give access to the site, a part doesn’t arrive. If moving a job means opening a form and filling in fields, it ends up not getting moved: it gets jotted down in a notebook and the system stops reflecting reality, which is the worst possible outcome.

That’s why it matters that rescheduling is just dragging the task to another day, both in desktop planning and from the technician’s own app, which has its own Gantt calendar view. The design principle is simple: if updating the plan costs more than ignoring it, it gets ignored.

Planning against availability, not an ideal calendar

A pretty Gantt chart that places work on a holiday or during someone’s vacation week isn’t planning: it’s manufacturing a delay in advance.

In GMAO Cloud, planning relies on staff information that exists for this purpose: shifts with their technicians, work schedules with their hours, vacations with their types and holidays. Before generating a preventive order, the system checks whether that day is a holiday and whether the technician is available.

Worth clarifying the scope so as not to create the wrong expectation: this isn’t HR software —no payroll, no contracts, no performance reviews. It’s the staffing part that maintenance needs in order to plan.

Teams, not just individuals

Once headcount passes ten people, planning at the individual level stops scaling. Technicians get grouped into teams with their own lead, so work can be assigned to the team and each lead sees their own without having the whole company’s orders in front of them.

And it works the other way too: the technician can see unassigned orders and take them themselves. Whoever’s on the ground knows better than anyone whether it’s on their way and whether they’re carrying the material. That takes the load of distributing work all day off whoever’s planning.

What the Gantt chart doesn’t know

There are two limitations worth being clear about, because they explain why a Gantt chart doesn’t replace judgment.

It doesn’t know the real duration until you measure it. A bar drawn with an estimated duration is a hypothesis. Only once times are logged with a stopwatch from within the order itself —and the order stores the target time, the estimate and the deviation— can you plan with actual durations. Until then, the plan will be optimistic, always in the same direction.

It doesn’t know what’s urgent. Who gets attended to first is decided by a person with the context in front of them: which client has a committed deadline, which piece of equipment is stopping production, which check can’t be skipped another month. The chart shows the consequence of the decision, not the decision itself.

Project Gantt and resource Gantt

Worth distinguishing two uses that look alike and aren’t the same, because mixing them up is the most common reason a schedule doesn’t work.

A project Gantt orders tasks that depend on each other: this can’t start until that finishes. It’s what’s used on-site and for new installations, and in GMAO Cloud it has its place in projects and construction jobs, where work is spread between two dates.

A resource Gantt —the subject of this article— orders people: who does what and when, with no dependencies between tasks. It’s the one that fits day-to-day maintenance, where jobs don’t depend on each other but compete for the same hands.

The usual confusion is asking the second one for things that belong to the first. A maintenance schedule doesn’t need dependencies between tasks: it needs to know whether that person has room for it.

How to know if it’s working

Three figures, all coming from reports and all dependent on times being logged as they happen.

Hours per technician, to see whether the distribution is real or just looks balanced on paper.

Deviation between estimated and actual time, which is what corrects the plan itself: if a type of job is systematically underestimated, next month’s bars will be wrong from day one.

Percentage of the annual preventive plan executed, which is the honest check of whether the schedule is being followed or just drawn.

Where to start

With the bare minimum needed for the plan to be credible: teams, work schedules and holidays loaded, and times being logged in the field. With that, the chart goes from being a drawing to being a decision-making tool.

If you’d like to see the scheduling with your own structure and your own frequencies, you can request a demo.