Every tag in a SCADA database has a point type, and the most basic division is digital versus analog. A digital point carries a state - on or off, open or closed, running or stopped - while an analog point carries a measured number that can take any value in a range. This classification shapes how each point is stored, displayed, alarmed, and reported, and it extends into counters and control points too. This page defines the SCADA point-type taxonomy from the data-modeling side, distinct from the hardware view of I/O.
Digital vs Analog Points in one line: A digital SCADA point represents a discrete state with a limited set of values - typically two, such as on and off or open and closed - and is used for status indications and alarms. An analog SCADA point represents a continuous measured value, such as a pressure or level, that can be any number within a range. Beyond these two core types, SCADA point taxonomies usually add counters, which accumulate totals, and control points, which send commands, so the full classification covers status, measured value, accumulated total, and command.
A digital point, also called a status or binary point, holds a discrete condition. In its simplest form it has two states - a pump is running or stopped, a valve is open or closed, an alarm contact is made or clear. The value is not a measurement to be scaled but a state to be interpreted, so a digital point is typically presented as a label, a color, or an icon rather than a number, and its alarming is about being in the wrong state rather than crossing a limit. Some digital points carry more than two states, but the essence is a small, enumerated set of conditions.
An analog point holds a continuously variable measured value. A pressure, a temperature, a tank level, a flow rate - each is a number that can sit anywhere within an engineering range and moves smoothly as the process changes. Analog points are scaled into engineering units, trended as curves, and alarmed against limits with deadbands, because the interesting questions about them are how high, how low, and how fast they are changing. The whole apparatus of thresholds, deadband, and rate-of-change alarming is built around this continuous nature.
This digital-versus-analog split is a modeling distinction, and it is worth separating from the hardware I/O it often mirrors. A discrete input from a field device usually becomes a digital point, and an analog input usually becomes an analog point, but the point type is about how the value is treated in the SCADA database - how it is stored, displayed, alarmed, and reported - not about the wiring. A calculated tag with no wiring at all still has a point type, and it is chosen on the same basis: does this value represent a state or a measurement?
Two more point types round out the common taxonomy. A counter, or accumulator, represents a running total rather than an instantaneous value - a volume totalizer, a run-hour meter, a pulse count from a flow meter. Counters behave differently from analog points because their natural behavior is to only increase, and the interesting quantity is often the difference between two readings over an interval, not the raw total. They need special handling for rollover, when the counter reaches its maximum and wraps back to zero, so the increment is computed correctly across the wrap.
Control points are the output side of the model - tags that send commands to the field rather than reporting from it. A digital control point issues a discrete command such as open, close, start, or stop, while an analog control point writes a setpoint or a variable output. Control points carry the extra machinery of safe command handling, such as select-before-operate and readback verification, because writing to the field has consequences that reading never does. In many taxonomies these mirror the input types, giving a symmetric picture of status and measured value coming in, command and setpoint going out.
Together these four families - status, measured value, accumulated total, and command - form the vocabulary a tag database is built from. Every point an engineer configures is classified into one of them, and that classification determines the properties the point needs, the way it is displayed, and how it can be alarmed or trended. Getting the type right at configuration time is what makes a totalizer roll over correctly, a status change into an alarm, and a control point behave safely, so the taxonomy is not academic - it is the foundation the whole data model rests on.
When a remote oil and gas site is modeled in cloud SCADA, its tag database is built by assigning each value the right point type. A wellhead's pump status and valve positions become digital points, its pressures and temperatures become analog points, its flow totalizers become counters, and any remote-operated valves or setpoints become control points. This classification is the first modeling decision, because everything downstream - the display, the alarm configuration, the daily report - depends on the type being correct.
A platform such as Merobix uses point type to apply the right behavior automatically: analog points get engineering-unit scaling, trends, and limit alarms with deadband; digital points get state labels and status alarms; counters get rollover-aware totalization; and control points get safe command handling. Because the type is set once when the point is defined, a site added later inherits consistent treatment, and an analog pressure at one well behaves the same way as an analog pressure at every other well. That consistency is what lets a fleet of sites be managed with one coherent data model.
The taxonomy also clarifies what can and cannot be asked of a point. You do not put a deadband on a pump-running status or a limit alarm on a valve command, and you do not trend a totalizer the way you trend an instantaneous pressure. Knowing the point type tells an engineer immediately which properties are relevant and which are meaningless, which speeds configuration and prevents the mistakes - like alarming a counter's raw total across a rollover - that come from treating a value as the wrong type.
A digital point represents a discrete state with a small set of values, such as a pump running or stopped or a valve open or closed, and is shown as a label or status. An analog point represents a continuous measured value, such as a pressure or level, that can be any number in a range and is scaled, trended, and alarmed against limits. The difference is state versus measurement.
They are closely related but not identical. Discrete I/O describes the hardware signal - an on/off input or output on a device - while a digital point describes how that state is modeled in the SCADA database, including how it is stored, displayed, and alarmed. A discrete input usually becomes a digital point, but the point type is a data-modeling concept, and even a calculated tag with no wiring has a point type.
The common taxonomy has four families: status or digital points for discrete states, analog points for continuous measured values, counters or accumulators for running totals, and control points for commands and setpoints sent to the field. Each type carries different properties and behaviors, so classifying a value correctly determines how it is scaled, displayed, alarmed, and reported.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.