What is process automation?

Process automation is the automatic measurement, control and monitoring of industrial processes in which materials are transformed, such as mixing, heating, reacting, separating and purifying, using control systems that keep process values like temperature, pressure, flow and level at their desired value without constant manual intervention. It is the backbone of the process industries: chemicals, refining, pharmaceuticals, food, energy and the water sector. Where machine automation drives the movements and steps of machines, process automation revolves around thousands of control loops that keep a physical or chemical process stable day and night.

By automating processes, companies increase their efficiency, quality, safety and reliability, while reducing costs and energy consumption.


๐Ÿง  How does process automation work?

The foundation is the closed feedback loop: measure, compare, correct. Process automation consists of layers and components that work together within a control system:

  1. Sensors and transmitters measure physical quantities such as temperature, pressure, flow, level and pH, and pass them on as a 4-20 mA signal or digitally.
  2. PLCs or DCS controllers collect the measurements, compare them with the desired values (setpoints) and use a PID algorithm to calculate a control output.
  3. Actuators such as control valves, pumps and motors with variable frequency drives carry out the correction.
  4. SCADA or HMI systems give operators in the control room an overview, alarms and the means to intervene.
  5. Higher levels such as a historian, MES and ERP analyse the data and link production to planning and quality.

This layering follows the automation pyramid and the Purdue Model. Alongside the normal control system (the Basic Process Control System), a hazardous plant has a separate safety instrumented system that brings the process to a safe state when control fails. On top of the PID loops, Advanced Process Control, often using model predictive control, optimises several variables at once.


๐Ÿ”„ What is the difference between process and machine automation?

Process automation controls continuous or batch flows of material; machine automation drives the discrete steps of machines. The distinction between discrete, batch and continuous processes determines which kind of control fits.

Aspect Process automation Machine automation
Type of process Continuous or batch processes Discrete processes
Aim Hold process values steady at setpoint Drive the movements and sequences of machines
Control strategies PID, cascade control, feedforward, model-based Logic, sequence control, motion control
Typical controller DCS, sometimes PLC with SCADA PLC, motion controller, robot controller
Time scale Seconds to hours Milliseconds
Downtime Start-up can take hours to days Often restarted within minutes
System lifetime 20 years or more 10 to 15 years
Examples Reactor, mixing tank, distillation column Packaging machine, sorting system, robotic arm

Batch processes, as in pharmaceuticals and food, sit in between: they are structured with ISA-88 and often run on batch control systems.


๐Ÿ”ง Which systems and components make up process automation?

Layer Function Examples
Field instrumentation Measure and act Pressure and flow transmitters, analysers, control valves
Field communication Signals to the controller 4-20 mA, HART, Foundation Fieldbus, Profibus PA, Ethernet-APL
Process control Control loops and sequences DCS such as PCS 7, PLCs programmed to IEC 61131
Safety layer Emergency shutdown and safe state SIS to IEC 61511
Operation Visualisation and alarms Operator stations, alarm management
Data and integration History, analysis and links Historian, OPC UA, MES

The design of a plant is captured in a P&ID, in which every measurement, control loop and valve gets its own tag number. In areas with an explosion hazard, ATEX requirements also apply, and field instruments are often supplied in an intrinsically safe design.


๐Ÿ•ฐ๏ธ How has process automation evolved?

Period Development
1940s Pneumatic controllers with a 3-15 psi signal; in 1942 Ziegler and Nichols publish their well-known tuning method for PID controllers
1950sโ€“1970s Electronic controllers; the 4-20 mA current loop becomes the standard for analogue signals
1975 Honeywell (TDC 2000) and Yokogawa (CENTUM) introduce the first distributed control systems
1980sโ€“1990s Smart transmitters with HART, followed by digital fieldbuses
Late 1990sโ€“2000s Ethernet and Windows in the control room, links to office networks
From around 2020 Ethernet-APL, Module Type Package, NAMUR Open Architecture and Open Process Automation

The 4 mA lower limit is a live zero: a current of 0 mA means a broken cable, not a reading of zero. Under NAMUR NE 43, a current of 3.6 mA or less, or 21 mA or more, signals a fault. The current trend is openness: modular plants connected plug-and-produce through MTP, and control systems that are no longer tied to a single DCS vendor.


๐Ÿญ Where is process automation used?

  • Chemical industry - precise dosing, pressure and temperature control of reactors
  • Refining and petrochemicals - distillation columns and crackers, often with APC
  • Pharmaceutical industry - batch production, validation and GMP-compliant process management
  • Food industry - pasteurisation, fermentation and hygienic cleaning (CIP)
  • Water sector - drinking water treatment, pumping stations and wastewater treatment with continuous measurement of flow and quality
  • Energy production - boilers, turbines and combined heat and power plants

Automation makes it possible to run these processes reliably, 24/7. In the Netherlands many of these plants are critical infrastructure, and large chemical sites fall under the Seveso rules: see Seveso and Cybersecurity and Water Sector OT Security.


๐Ÿงญ How do you approach a process automation project?

A DCS or process control system lasts for decades, so the decisions made at the start carry a lot of weight. A common approach:

  1. Process design and P&ID - define what needs to be measured and controlled.
  2. Risk analysis - a HAZOP and LOPA determine which safety functions are needed and which SIL they must achieve.
  3. Functional specification - control strategies, interlocks, alarms and operator displays, preferably following ISA-101 for the HMI.
  4. Security by design - zones and conduits according to IEC 62443 and a security risk assessment of the control system and SIS.
  5. Engineering and testing - configuration, a factory acceptance test (FAT) and then a site acceptance test (SAT).
  6. Commissioning and tuning - tune the control loops and rationalise the alarm system.
  7. Operation and maintenance - change management, backups, patches and periodic proof tests of the SIS.

When tuning, it helps to know what dynamics to expect:

Control loop Typical response time Note
Flow Seconds Fast and often somewhat noisy; usually PI control
Pressure (gas) Seconds to minutes Depends on the volume of the system
Level Minutes Often deliberately tuned slowly to smooth the flow downstream
Temperature Minutes to hours Long lag; often cascade control onto a flow loop
Composition (analyser) Minutes to hours Measurement delay makes APC or feedforward worthwhile

For alarms, the state of the art is set out in ISA-18.2 and EEMUA 191. Under ISA-18.2, an alarm flood begins as soon as an operator receives more than ten alarms within ten minutes.


๐Ÿ” How do you secure process automation?

Process control systems are increasingly connected to IT systems, vendors and the cloud. An attack can therefore have physical consequences: Stuxnet, discovered in 2010, sabotaged Iranian uranium centrifuges through their control system, and TRITON targeted the safety system of a petrochemical plant in 2017. Key measures:

IEC 62443 is the international series of standards for the security of industrial automation and control systems.

Many process companies, such as drinking water utilities, chemical plants and food producers, fall under NIS2. In the Netherlands that directive has been transposed into the Cybersecurity Act, which entered into force on 15 August 2026.


โ“ Frequently asked questions

What is the difference between process automation and industrial automation?

Industrial automation is the umbrella term for all automation in factories and plants. Process automation is the part that deals with continuous and batch processes involving flows of material, as in chemicals, water and food. The other part, machine or factory automation, controls discrete manufacturing.

Is process automation done with a PLC or a DCS?

Process automation can use either. Large continuous plants with thousands of control loops, such as refineries, usually use a DCS, because control, operation and engineering form one integrated and redundant whole. Smaller or geographically spread installations, such as pumping stations and treatment plants, often run on PLCs with SCADA.

What is a control loop in process automation?

A control loop is the smallest building block of process automation: a sensor measures a process value, a controller compares it with the setpoint and an actuator such as a control valve makes the correction. The controller usually runs a PID algorithm. A medium-sized plant has hundreds to thousands of these control loops.

Which standards matter for process automation?

The most important standards for process automation are ISA-88 (batch control), ISA-95 (integration with MES and ERP), ISA-18.2 (alarm management), IEC 61511 (safety instrumented systems) and IEC 62443 (cybersecurity). In addition, NAMUR, the association of users of process automation technology, publishes widely used recommendations such as NE 43 and NE 107.

Why is process automation vulnerable to cyberattacks?

Process automation systems often run for decades, cannot simply be restarted for updates and use older protocols without authentication. At the same time they are increasingly connected to IT networks and remote vendors. An attacker who reaches the control system or the SIS can disrupt the physical process, with consequences for safety and the environment.

What are the benefits of process automation?

Process automation delivers more consistent product quality, lower energy and raw material use, greater safety and less manual work. Because control loops correct continuously, a plant can run closer to its optimal operating point. Data from process automation also forms the basis for optimisation, scalable expansion and predictive maintenance.


๐Ÿ“Œ In summary

Process automation uses sensors, control loops, a DCS or PLC and a separate safety system to keep continuous and batch processes stable, efficient and safe, day and night. It provides scalability, repeatability and reliability, from simple pump control to a fully automated refinery. Because it is connected to IT networks, segmentation and security according to IEC 62443 and IEC 61511 belong in the design from day one.