Best Practices for Designing Custom RFID Systems

A good RFID system should be designed around the way your operation actually works.

That sounds obvious, but it is easy to get backwards. A company starts looking at RFID readers, tags or software before clearly defining what needs to improve. Then the technology starts driving the process instead of supporting it.

Metalcraft and ARK Business Systems start with the application, not the technology. We look at what you’re trying to track, how it moves through your operation, where you need better visibility and what systems are already in place. From there, we determine what combination of tags, hardware, software and integration makes the most sense for your operation. That may mean a relatively simple RFID asset tracking system using handheld readers and existing software. It may involve fixed readers, custom software and ERP integration. In some cases, RFID may not even be the best answer for every part of the workflow.

The point is to understand the problem first and then determine what combination of identification, hardware, software and integration makes sense.

Why Custom RFID Systems Are Sometimes Necessary

There are a lot of good off-the-shelf RFID products available. For many applications, that is exactly what you should use. If you have a straightforward asset tracking need and a packaged system can handle it, there is no reason to make the project more complicated than it needs to be.

Custom RFID system design becomes more important when the workflow, environment or integration requirements are outside the norm.

For example:

• Assets move through several departments and need to trigger different actions along the way.
• RFID tags need to perform on metal, near liquids or in another difficult environment.
• A standard software package does not match the way employees actually work.
• RFID data needs to connect with ERP, WMS, MES, CMMS or another existing system.
• Fixed readers need to capture movement automatically without reading unrelated assets nearby.
• Several different asset types require different tag constructions.
• The operation needs mobile workflows that do not exist in standard software.

The more variables an application has, the more important it becomes to look at the system as a whole. Tag selection affects read range. The way an asset moves affects reader and antenna placement. The physical environment can change RFID performance. And even if the hardware works exactly as expected, the system still needs to collect the right data and get it into the right place.

That doesn’t necessarily mean the system needs to be complicated or completely custom. In some cases, a standard RFID tag, handheld reader and packaged software may be all that is needed. In others, the application may require specialized tags, fixed readers, multiple antennas, custom software or integration with existing systems.

The key is figuring that out before selecting the technology. That starts with understanding the application and defining exactly what the system needs to accomplish.

Phase 01 — Understand the Application

Before deciding on RFID tags, readers or antennas, define what the system needs to accomplish.

This usually starts with a few basic questions:

• What are you trying to track?
• Where does it move?
• Where do you lose visibility today?
• What part of the process is manual?
• What information do employees need?
• Where should that information go?
• What would success look like?

The answers sound simple, but they drive almost every technical decision that comes later. If the objective is faster inventory counts, the design may center on handheld UHF RFID readers. If the goal is automatic movement tracking through a production area, fixed readers and defined read zones may make more sense. If the problem is finding equipment across multiple locations, the software and data model may matter just as much as the RFID hardware. The first phase should also document the existing environment. That includes the physical assets as well as the systems already in place.

Asset characteristics to record include:

• Material
• Size and shape
• Available tagging space
• Value
• Mobility
• Environmental exposure
• Current identification
• Required read distance

Existing systems may include:

• ERP
• WMS
• MES
• CMMS
• Inventory management software
• Asset management platforms
• Custom databases
• Mobile applications

It’s also important to define what success looks like before designing the system. Simply saying you want to implement RFID doesn’t provide much direction. A better goal might be reducing quarterly inventory from two days to four hours, automatically tracking when a tool leaves the crib or eliminating manual data entry between production and your ERP. Having a clear outcome in mind helps determine how the system should be designed and gives you something measurable to test against.

Build the RFID Tagging Strategy Around the Asset

RFID tags should be selected by asset class, not by convenience. A tag that works on cardboard may not work on metal. A large RFID label may have excellent read range but be impossible to fit on the asset. A standard adhesive may work in an office and fail quickly in a washdown environment.

That means the RFID tagging strategy needs to account for:

• Surface material
• Available space
• Attachment method
• Read range
• Tag orientation
• Environmental exposure
• Expected asset life
• Data and encoding requirements

Metal surfaces are one of the most common examples. Metal reflects radio waves and can interfere with ordinary RFID tags. In those applications, a tag specifically designed for metal such as Metalcraft’s Universal Mini RFID Tag may be required. Liquids create a different problem because they can absorb RF energy and reduce UHF RFID performance. The right tag depends on the actual asset and environment.

Attachment matters too. Depending on the application, tags may use:

• Adhesive
• Screws
• Rivets
• Cable ties
• Embedded mounting
• Other mechanical attachment

The environment also plays a role in determining the right tag construction. If the tag will be exposed to chemicals, abrasion, impact or extreme temperatures, it needs to be designed to withstand those conditions.

You also need to consider what information the tag needs to contain. That may be a unique identifier, EPC, serial number or asset ID, along with a barcode and human-readable information when needed. The goal isn’t to put as much information as possible on the tag. It’s to include what is needed to identify the asset and support the workflow, while keeping more detailed information in the software or existing system when it makes sense.

Plan Reader and Antenna Coverage Around the Workflow

The next question is where reads need to happen.

That may include:

• Receiving
• Warehouse aisles
• Tool rooms
• Production cells
• Doorways
• Maintenance areas
• Shipping
• Vehicles

The type of reader you need also depends on how assets move through your operation. Handheld readers make sense for things like inventory counts and locating assets, while fixed readers can automatically capture assets as they move through specific areas. For larger warehouses, distribution centers or outdoor yards, mobile or vehicle-mounted readers may be a better fit.

Reader selection is only part of the equation. Where readers and antennas are placed—and how the antennas are positioned—can have a significant impact on read range and accuracy. Depending on the application, multiple antennas may be needed to account for different asset and tag orientations.

And longer read range isn’t always better. The goal is to consistently read the right tag in the right place without picking up tags from areas you don’t want to include. That’s why reader and antenna placement needs to be designed around the actual workflow and tested in the environment where the system will be used.

Phase 02 — Design the RFID Topology

Once you understand the requirements, the next step is figuring out how the system will work in the actual environment. This is especially important with RFID because the surrounding environment can have a big impact on performance. Metal can reflect RFID signals, liquids can absorb them and machinery can introduce interference. Even walls, doors and shelving can affect where and how consistently tags are read.

That’s why reader and antenna placement can’t be based on a floor plan alone. You need to understand how RFID will perform in the areas where reads need to happen and account for those conditions when designing the system.

The RFID frequency also needs to match the application. Low-frequency RFID is typically used for shorter-range applications, while high-frequency RFID at 13.56 MHz works well for controlled, close-range interactions such as access control. Ultra-high-frequency RFID, including RAIN RFID, is commonly used for applications like asset tracking, inventory and supply chain visibility where longer read ranges or the ability to read multiple tags are important.

But just because you can read a tag from farther away doesn’t mean you should. A 30-foot read range isn’t much of a benefit if you’re also picking up inventory from the next aisle. The goal is to create reliable read zones that capture the tags you want without capturing the ones you don’t.

Test the Design on the Floor

A site survey is an important part of designing an RFID system because it gives you the opportunity to test the system in the environment where it will actually be used. A floor plan can help determine where readers and antennas might be placed, but onsite testing is what tells you whether those locations will deliver the performance you need.

Whenever possible, testing should be done using the actual or representative assets, tags, readers and antennas so you can see how the complete system performs under real operating conditions.

Look at:

• Steel shelving
• Machinery
• Liquids
• Walls
• Doors
• Other RFID equipment
• Employee traffic
• Storage patterns
• Nearby electrical equipment

Then collect actual read-rate data.

If results are inconsistent, adjust:

• Reader location
• Antenna angle
• Antenna height
• Reader power
• Number of antennas
• Tag placement

From there, it’s a matter of testing and making adjustments until the system performs the way it needs to. That may mean changing reader power, moving an antenna or adjusting its orientation and then testing again. It’s not unusual to make several adjustments during this process. That’s the purpose of testing—to understand how the hardware performs in the actual environment and fine-tune the system before it is fully deployed.

Phase 03 — Turn the Design Into a Build-Ready Specification

Once testing is complete and you know what the system needs to look like, the next step is documenting everything needed for implementation. This should clearly define the hardware, how all of the components work together and what needs to happen with the data once it is captured.

The bill of materials may include:

• RFID tags
• Readers
• Antennas
• Cables
• Mounting hardware
• Printers
• Handheld devices
• Network hardware
• Power supplies
• Enclosures

The specification should also include acceptable alternatives when appropriate so the project isn’t dependent on a single piece of hardware.

The same level of detail is needed for the software and data. RFID readers can capture a lot of information very quickly, but not every read needs to become a transaction in your business system. For example, a tag sitting in a read zone for 20 seconds may be read multiple times when the system really only needs to know that the asset entered or left that area.

The software needs to take those individual reads and turn them into useful information that supports the workflow.

For example:

• Asset entered Room A.
• Tool left the crib.
• Container arrived at shipping.
• Inventory item was detected at Location 4.

The integration design should define:

• IDs
• Timestamps
• Locations
• Event types
• APIs
• Error handling
• Authentication
• Data ownership

This is an important part of getting RFID implementation right. Even if the tags, readers and antennas are working exactly as expected, the system won’t provide much value if the data doesn’t support the workflow. Defining how that information will be used before implementation helps make sure all the pieces work together once the system is deployed.

Testing, Read-Rate Tuning and Acceptance

Testing should always go back to what you wanted the RFID system to accomplish in the first place. If the goal is to read 99% of the assets moving through a portal, testing should recreate that workflow and measure the actual read rate. If the goal is to reduce inventory time to less than 20 minutes, then that’s what you need to test. This helps make sure you’re measuring the performance of the system based on the results that actually matter to your operation.

Testing should include:

• Successful reads
• Missed reads
• False reads
• Read distance
• Tag orientation
• Multiple-tag performance
• System response
• Integration accuracy

Testing also needs to reflect the number of tags the system will encounter during normal use. Successfully reading one tag doesn’t necessarily mean the system will perform the same way when dozens or even hundreds of tags are in the read zone at the same time.

If testing shows the system isn’t meeting the requirements, adjustments can be made to the antenna orientation, reader power, antenna location or even the tag placement. From there, test again and continue making adjustments until you get the performance you need. Any changes made during testing should also be reflected in the final system design and bill of materials.

What Should Be Delivered After Field Testing?

Once testing is complete, the final documentation should reflect the installed system rather than the original concept.

That may include:

• Final reader locations
• Antenna coverage maps
• Cable routing
• Network design
• Final hardware BOM
• Reader settings
• Test logs
• Read-rate results
• Integration results

This becomes the baseline for maintenance and future expansion.

Deployment and Training

Depending on the size of the RFID implementation, it may make sense to roll it out in phases. Starting with one area, asset class or workflow gives you the opportunity to make sure the system is working as expected and employees are comfortable with the new process before expanding it to other areas. Training should focus not only on how to use the technology, but also on how it fits into the workflow and what employees will need to do differently.

Employees need to know:

• How assets are tagged
• How RFID reads occur
• What a successful transaction looks like
• What to do when something does not read
• How exceptions are handled
• Who owns data corrections

At the end of the day, the RFID system should make the process easier, not add more work. If employees have to add several new steps just to eliminate one manual task, it’s worth taking another look at the workflow. Once the system is in place, continue tracking the same goals and metrics that drove the project in the first place to make sure it is delivering the improvements you expected.

That may include:

• Inventory time
• Inventory accuracy
• Search time
• Asset utilization
• Missed reads
• Exception rates
• Labor savings
• Throughput

Those numbers tell you whether the RFID system is actually improving operational efficiency.

How Metalcraft and ARK Fit Into the System

Metalcraft and ARK Business Systems bring together the different areas of expertise needed to develop a complete RFID solution. Depending on the application, that may include tag and inlay design, readers and antennas, software, data management, integration, testing and implementation.

Having all of those areas working together is important because an issue with one part of the system can affect another. A read-rate problem could be related to the tag, where it is placed, the surface it is attached to, reader and antenna placement or something in the surrounding environment. What appears to be a software problem may actually be related to the workflow or the way data is being captured.

Looking at the entire application as one system makes it easier to determine what is really happening, make the right adjustments and keep the focus on the outcome the customer is trying to achieve.

Asset Tracking, Inventory Tracking or Custom Software?

Not every RFID application requires custom software. In many cases, one of our existing platforms already provides the functionality needed. Grey Trunk RFID is a good fit for asset tracking applications where you need to identify, locate and manage equipment across one or multiple locations. QuickTrack RFID is designed more around inventory management, including receiving, cycle counts and inventory visibility.

When the workflow or requirements don’t fit within either platform, that’s when it makes sense to look at a more customized approach.

That may happen when:

• The workflow is unique.
• RFID data needs to trigger custom actions.
• ERP, MES, WMS or another existing system requires deeper integration.
• Employees need a purpose-built mobile workflow.
• Standard software creates too many workarounds.

That is where custom development makes sense. The point is not to build custom software whenever possible. It is to avoid building it when it is unnecessary—and use it when the workflow genuinely requires it.

Security and Compliance Need to Be Part of the Design

Security also needs to be considered early in the design process, especially when the RFID system will be collecting or connecting to sensitive information. Depending on the application, that could include operational data, employee or location information and, in healthcare applications, patient or clinical information.

Depending on the application, the design may need to account for:

• SOC 2
• GDPR
• HIPAA
• Role-based access
• Authentication
• Encryption
• Audit logs
• Data retention
• API security

The RFID tag itself typically only needs to contain enough information to identify the asset. Any sensitive or more detailed information can be stored in the secured system and connected to the asset through its unique identifier. This keeps unnecessary information off the tag while giving you more control over who can access the data and how it is managed.

Designing RFID for Harsh Environments

The system also needs to survive where it is installed.

RFID tags and hardware may encounter:

• Metal surfaces
• Chemicals
• Moisture
• Outdoor exposure
• High or low temperatures
• Impact
• Vibration
• Dust

The hardware and tags also need to be selected based on the environment where they will actually be used. That may mean using a rugged enclosure to protect a fixed reader in a manufacturing environment, a chemical-resistant tag on production equipment or an on-metal RFID tag for assets with metal surfaces. Hazardous environments may also require equipment with specific certifications, such as ATEX or IECEx. Identifying these conditions early helps make sure the right components are included in the system design from the beginning.

What a Complete RFID System Design Should Include

By the end of the design process, the project team should have enough information to build, install and test the system without having to make major design decisions in the field.

That usually includes:

• RFID tagging specifications
• Hardware bill of materials
• Reader locations
• Antenna placement
• Network requirements
• Power requirements
• Software architecture
• Integration requirements
• Data models
• Acceptance criteria
• Test results
• Training documentation
• Maintenance information

Having all of this documented before implementation helps make sure everyone involved understands what the system should look like, how it should work and what is needed to get it up and running successfully.

Start With the Problem

A successful RFID project starts with understanding the problem you’re trying to solve, not deciding which tag, reader or software you want to use. Maybe inventory takes too long, employees spend too much time looking for tools, you don’t have enough visibility into where materials are in the process or a manual process is creating too many errors.

Once you understand what you want to improve, you can determine what technology makes the most sense. That may be an existing RFID platform, a custom solution or even a combination of RFID and barcode. The technology should fit the application, not the other way around.

Start with what you want to accomplish, and then determine what it will take to get you there.

Frequently Asked Questions

How Do You Create Your Own RFID System?

Start by defining the business problem, then document the assets, workflows, environment and existing systems involved. From there, select the appropriate RFID tags, readers, antennas and software, conduct a site survey and test the proposed configuration. A pilot should validate read performance and workflow improvements before the system is expanded.

What Are the Three Types of RFID?

RFID tags are commonly categorized as passive, active and semi-passive. Passive tags receive power from the RFID reader signal. Active tags contain their own power source and support longer-range communication. Semi-passive tags contain a battery for certain functions but typically rely on a reader to initiate communication. RFID is also categorized by LF, HF and UHF frequency.

How Much Does It Cost to Implement an RFID System?

The cost depends on the number of assets, RFID tags, readers, antennas, software, integrations, installation and testing involved. A handheld inventory application may require relatively little infrastructure, while a multi-facility fixed-reader system can require a larger upfront investment. A feasibility study and pilot provide a much more useful estimate than general industry averages.

What Are Some Examples of RFID Projects?

Common RFID projects include tool tracking, inventory management, fixed asset tracking, work-in-process, returnable container tracking, IT asset management, vehicle identification, access control and supply chain visibility. Custom RFID systems are most useful when standard hardware or software does not match the workflow, environment or integration requirements.