Considerations for choosing the right control system platform

Cover Story: Choosing between a programmable logic controller (PLC) or a distributed control system (DCS) depends on the type of project and plant. PLCs and a DCS provide different kinds of benefits, depending on the application.

By Bob Halgren January 9, 2017

The foundation for a successful automation project requires automation and design engineers to evaluate application requirements as well as choose the most effective control system platform. These decisions will have a long-term impact on a facility’s operational performance for as long as 25 years in some cases as long as the process is in place. Generally, most control system decisions come down to choosing a programmable logic controller (PLC) or a distributed control system (DCS). In some cases, one option is clearly better for a plant while in others the option is not as easy. There are many factors involved in selecting the control system that will help the user achieve its short- and long-term goals.

Control system platforms influence how the automation system will meet the needs for optimizing production, sustaining reliability, and obtaining data. However, a lack of foresight in choosing a control system can also lead to potential roadblocks for expansion, process optimization, user satisfaction, and a company’s profits.

Apart from basic criteria—such as how to control the process—the design team must recognize considerations such as installation, expendability, maintenance, and upkeep, among others.

While PLCs might be cost-effective for the time being for today’s small facility, a DCS provides a more economical expansion with a potential return on the initial investment. 

Differences between system platforms

A PLC is an industrial computer that is built to control manufacturing processes such as robots, high-speed packaging, bottling, and motion control. In the last 20 years, PLCs have gained functionality and provided benefits to small plant applications and skid units. PLCs are generally self-contained islands of automation that can be integrated so they can communicate with one another. The engineering required for integration requires some degree of mapping between controllers because each PLC has its own database. This makes PLCs a good fit for small applications that are unlikely to be expanded.

A DCS distributes controllers throughout the automation system and offers a common interface, advanced control, a systemwide database, and information that can be shared easily. DCSs are traditionally used in process applications and larger plants and are easier to maintain for large system applications throughout the plant’s lifecycle.

Application type determines the platform

PLCs and DCSs are generally suited to one of two manufacturing types: discrete and process manufacturing. Discrete manufacturing facilities, which usually use PLCs, are composed of separate production units that generally assemble components, such as labeling or fill-and-finish applications. Process manufacturing facilities, which usually use DCSs, automate continuous and batch processes and implement formulas composed on ingredients rather than pieces, and their output is measured in bulk. Large continuous process facilities, such as refineries and chemical plants, use DCS automation. Hybrid situations that are composed of both manufacturing types generally use a combination of PLCs and DCSs.

Project teams need to consider several aspects to determine which system better fits the application. While the aspects are listed separately here, in practice they are interwoven:

  • Process size
  • Expansion or modification plans
  • Integration needs
  • Functionality
  • High availability
  • Return on investment (ROI) over the facility lifecycle. 

Process size: How many input/output (I/O) points does the process require? Smaller systems (<300 I/O points) might have smaller budgets, which makes PLCs a better fit. DCSs don’t scale down as easily and function better in larger plants. They’re also easier to manage and upgrade because they have one database, and any changes that are made can be applied globally.

Modification plans: Small processes can use PLCs, but if that process is expanded or modified, more PLCs and databases need to be added and independently maintained. This is a time-consuming process that leads to errors. DCS upgrades are easier to perform, and aspects such as user credentials are managed from a central hub, which results in easier upkeep and maintenance (see Figure 1). 

Integration needs: For a skid that stands alone, a PLC is ideal. When multiple PLCs are brought into a plant, interconnections are required. These can be difficult to create and often require data mapping through communications protocols. Integration can be achieved, but users might run into challenges when changes are made to a PLC that can result in two PLCs that no longer communicate as intended because the data maps have been impacted. Mapping is not required with a DCS, and configuration changes are a simple process because the controllers are native to the system. 

Functionality: Some industries and facilities require historians, streamlined alarm management, and a central control room with common user interfaces. Some also require management execution system (MES) integration, advanced control, and asset management. The DCS works best for this situation because it has these applications built in (see Figure 2), which makes it easy to add on to an automation project without the need for separate servers or added integration costs. This is more cost-effective, increases productivity, and lowers risk. 

High availability: Consider the need for high control system availability-what are the risks and consequences of an unexpected controller failure? If users can accept a failure without a major production or safety impact, the PLC works well. If high process availability is required, the DCS provides redundancy options that are straightforward and cost effective (see Figure 3). 

Lifecycle ROI: Facility expectations vary across industries. PLCs provide excellent ROI for small and nonexpanding processes that don’t require integration with other process areas. While a DCS may have high installation costs, the full lifecycle costs and resulting increases in production and safety pay for themselves over the course of the system’s lifetime. 

Automation strategy

Balancing short-term needs with long-term vision is critical for operational certainty. When automation strategy is aligned with your business needs, the platform-whether PLC or DCS-satisfies criteria such as integration needs and required functionality. The values supplied by each platform will improve how the plant is operated and maintained.

Bob Halgren, DCS strategy manager, Emerson. Edited by Chris Vavra, production editor, Control Engineering, CFE Media, cvavra@cfemedia.com.

MORE ADVICE

Key concepts

Programmable logic controllers (PLCs) are mainly designed for discrete industries, and distributed control systems (DCSs) are mainly designed for process industries.

PLCs are best used for short-term applications while DCSs are best in long-term applications.

Several ways can determine if a PLC or a DCS is best for an operation.

Consider this

What steps do you use to determine whether to use a DCS or PLC for an operation?

ONLINE extra

See related articles on DCSs and PLCs linked below.


Author Bio: Bob Halgren, DeltaV platform management director, Emerson.