Some conditions are not captured by any single number but by how several numbers sit together, and watching half a dozen separate trends makes that overall pattern hard to see. A radar chart addresses this by drawing several related variables on axes that radiate from a common centre, so the values join into a single shape whose outline is the signature of the current state. This guide explains what a radar chart is, how a normal operating shape is compared against the current one, and when the format genuinely helps versus when it can mislead.
Radar chart in one line: A radar chart on an HMI, also called a spider or star chart, plots several related variables on separate axes radiating from a central point and joins the points into a closed shape. Because each state produces a characteristic outline, an operator can compare the current shape against a known normal shape to see multi-parameter conditions such as equipment health at a glance, though the format can mislead for precise reading or for unrelated variables.
A radar chart takes a set of related variables and gives each its own axis, with all the axes radiating out from a shared centre like spokes of a wheel. Each variable's value is marked as a distance from the centre along its axis, near the centre for a low value and far out for a high one, and the marked points are then joined by lines to form a closed polygon. That polygon, the shape traced out by all the variables together, is the essence of the chart, because it turns a collection of separate readings into a single visual object the eye can grasp as a whole.
The reason to do this is that the shape carries information the individual numbers do not make obvious. When several parameters describe one thing, such as the health of a machine seen through its vibration, temperature, load, and speed, their combined pattern says more than any one of them alone, and the radar chart makes that pattern into a recognisable form. A balanced, roughly regular polygon means the parameters are all in proportion; a shape pulled out on one axis or dented in on another shows that something is out of its usual balance, which is a multivariate observation a set of separate gauges does not surface.
Because the format is about pattern rather than precision, it is used where the relationship among several variables matters more than the exact value of each. It is common for multi-parameter equipment health, for comparing a small set of related quality or performance metrics, and for any situation where the operator's real question is whether the overall profile looks right, not what the seventh decimal of one variable is. The chart answers that shape-level question directly.
The radar chart becomes most useful when the current shape is compared against a reference shape that represents normal, healthy operation. A known-good profile, gathered when the equipment or process was running well, is drawn as one polygon, and the live values are drawn as a second polygon over the same axes. When everything is normal the two shapes overlap closely; when a parameter drifts, the live shape bulges or pinches away from the reference on that axis, and the operator sees the departure as a visible gap between the two outlines rather than as a number they must judge against a remembered baseline.
This comparison is the real strength of the format, because it converts abnormality into a difference in shape, which people notice easily and pre-attentively. An operator does not need to remember what normal looks like on six separate trends; they simply see whether the live outline matches the reference outline, and where it does not. The direction and size of the deviation on each axis also give a clue to what is wrong, since a shape that has grown along the vibration axis while the rest holds steady points to a different problem than one that has grown along the temperature axis.
Some implementations extend this with more than one reference shape, such as a normal envelope and a warning envelope, so the live polygon can be seen crossing from one region into another. Others overlay several items at once, comparing the shape of one machine against another to spot the odd one out. In every case the underlying idea is the same: the meaning lives in how the current shape relates to a reference, and the chart is designed so that relationship is read as geometry.
Radar charts are genuinely helpful when their preconditions hold: the variables are related and belong together, there is a meaningful normal shape to compare against, and the question is about the overall multivariate pattern rather than a precise single value. Under those conditions the chart does something separate trends cannot, giving an at-a-glance signature of a multi-parameter state, which is why it shows up in equipment-health and condition-monitoring displays where several sensors describe one asset. Kept to a modest number of axes and used for comparison, it earns its place.
It misleads when those conditions are not met. Reading an exact value off a radar axis is hard and error-prone, so the format is poor when precision matters. The shape also depends heavily on how the axes are ordered and scaled, since reordering the same variables produces a completely different-looking polygon, which means the shape can imply relationships that are artefacts of the layout rather than the data. Cramming in too many axes makes the shape a cluttered star that is hard to read, and using it for variables that are not actually related invites the eye to see a meaningful shape where there is none. For those cases separate trends or bars are clearer.
In a SCADA context, a cloud platform such as Merobix supplies exactly the multi-parameter, historical data that makes radar comparison possible, because it collects the several readings that describe an asset and retains the history from which a normal profile can be established. A condition-monitoring screen for a remote pump or compressor can draw the asset's current multi-sensor shape against its own known-good shape, letting an operator watching many sites spot the one whose profile has distorted. The sensible practice is to reserve the radar chart for that specific job, a small set of related parameters compared against a baseline, and to keep using ordinary trends and bars for precise single values and for variables that do not form a natural group, so the format's strengths are used and its weaknesses avoided.
A radar chart is also known as a spider chart or a star chart, and sometimes a polar or star plot, because the axes radiate from a central point like spokes or a star and the joined values trace a web-like shape. The names all refer to the same format: several variables plotted on axes around a common centre and connected into a closed polygon. It is used to show a multivariate profile as a single shape.
Use a radar chart when several related variables together describe one thing, such as an asset's health, and the question is whether the overall profile looks right rather than what any single value is precisely. It is especially useful when you can compare the current shape against a known normal shape. For precise single values, or for variables that are not genuinely related, separate trends or bars are clearer and less prone to misleading.
They are hard to read precisely, so they are poor when exact values matter. The apparent shape also depends on how the axes are ordered and scaled, so reordering the same data changes the polygon and can imply relationships that are artefacts of the layout. Too many axes make the shape cluttered, and using the format for unrelated variables invites the eye to see meaning that is not there. They work best for a small set of related variables compared against a baseline.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.