Do I need a thick or thin client for my Autocoding project?
Thin clients seem to offer many advantages over thick clients so why are thick clients still being used for some processes? We spoke to Wayne Johnson, head of the OAL Connected team, to understand the difference between thin clients and thick clients, and why thick clients are preferred for Autocoding applications.
What are thick and thin clients?
A ‘client’ refers to the type of PC unit used to run your automation equipment.
A thin client is a simple piece of hardware that looks like a small desktop PC but relies on the server for its processing power. This means it requires constant communication with the server to work.
The reduced amount of hardware inside thin clients makes them more secure than other options, cheaper to buy and run, and easier for your IT team to swap out and replace if a unit stops working. However, as the server handles the bulk of the workload, a thin client doesn’t have a hard disk drive for local data storage. If the connection to the server is lost, so is all the data as there is nowhere for the thin client to store it. There are also no connections for external devices such as printers, which may cause problems in certain applications.
A thick client is a traditional desktop-style PC tower unit. It is connected to the server but will process the data itself and transmit it to the server for archiving, rather than relying on the server to do the work.
Thick clients can cost more upfront, may take longer for your IT teams to deploy them and may be considered to have more security vulnerabilities than thin clients. However, you’ll need fewer servers with thick clients, which can reduce costs significantly. What’s more, as thick clients have their own hard disk drive, they can store production data locally. This means it only requires intermittent access to a server and can operate independently, as well as working uninterrupted, even during a power cut or an IT upgrade. They also have superior processing power compared to a thin client, which can support greater functionality on your line.
Why does OAL Autocoding use thick clients?
The Autocoding system is used to prevent product recalls and ensure that products are in the correct packaging. The risks of a customer getting an allergen-containing product that is incorrectly labelled can be life threatening for the customer, as well as costly for the manufacturer: product recalls are estimated to cost £150k or more when the cost of the fine, the ingredients and packaging wasted and the cost to reproduce are taken into account.
Because of this, it is critical that your Autocoding system is always working. With a thin client, if the network connection to the server is interrupted, the risk of a product not being verified is high. While networks are more robust than they used to be, wireless networks can still be intermittent and wired connections can be easily severed in a power cut or with a loose cable.
Even if your network connection is stable, there can be a risk of the connection to the server being slow as the information buffers. It’s possible that a mismatched label could have already gone down the production line before the mismatch is detected, making it harder to find the problem.
By using a thick client, verification can continue even if the network connection is lost or slow, as the thick client can use its local storage until the connection is restored. The savings that thin clients offer during installation can easily be wiped out with one product recall – is it worth the risk?
Another point to consider is that some retailer Codes of Practice will specify that a thick client is required for automated label verification processes. For example, Marks and Spencer is one retailer that specifies this. However, even if your retailer doesn’t specify a thick client, they can give you built-in resilience in your production line should a problem occur.
Finally, if verification speed is critical, you may be looking at using advanced systems such as artificial intelligence to reduce the time for quality checks. APRIL™ Eye, which uses artificial intelligence to read date codes on packages, can scan up to 1,000 packages per minute – but this requires much greater computing power than a thin client could provide.
Who uses thick clients?
Dawnfresh Seafoods had previously installed an Autocoding system, but it wasn’t fit for purpose and was creating significant production delays and downtime. It also didn’t comply with M&S’ Code of Practice; causing problems with the customer.
Dawnfresh turned to OAL for a solution to reduce their product recalls and ensure compliance. Because of the thick client architecture that we provide with our Autocoding system, Dawnfresh could achieve compliance with M&S’ Code of Practice while also delivering a 10% increase in packaging line throughput.
When should I use a thin client?
Thin clients offer many advantages and are better suited to some applications more than others, such as data capture processes like SCADA. At OAL, we believe that each system has a place in automation, and will help you choose the best solution for your needs.
About the author
Wayne Johnson has worked in the food industry for over 20 years, with experience on both sides of the retail fence as a technical manager for First Milk and a supermarket auditor for Somerfield. He now heads up OAL Connected helping manufacturers eliminate label and date code errors with our market-leading Autocoding system