Track and trace to avoid quality control disasters
Tuesday, 28 January, 2014
It is a critical aspect of quality assurance and food safety to identify a product at its origin and track and trace the forward movement through the entire process of the food supply chain, including food processing, filling, closure and packaging. To retrospectively identify where and when a problem may have occurred, and to date and time stamp the event, is a prerequisite.
The ability to automatically recognise defects up and down the supply chain has become an integral part of quality control in the food industry business and is a legal requirement in many processes.
Remarkably, the vast majority of food processing facilities around Australia still rely on human beings to do the track and trace inspections.
The problem is that human beings cannot, for the most part, keep pace with industry requirements for increased production speeds and increased product quality. Operators get tired, lose focus and make mistakes. This loss of concentration could result in them failing to detect potentially defective products; the criteria that they apply during inspections are inevitably subjective.
Machine vision (MV) is the technology to replace or complement manual inspections and measurements with digital cameras and image processing. The technology is used in a variety of different industries to automate the production processes, increase production speed and yield, and to improve product quality.
Typical applications of machine vision in the Australian market:
- Code validation - confirming that all the codes on the product are readable.
- Label inspection - ascertaining that the labels are placed correctly.
- Label validation - validating that the correct label is affixed to the product.
- Packaging integrity - ensuring that any recycled packaging material is adequate.
- Filling monitoring - validating that the correct product and correct amount is filled.
- Closure inspection - ensuring that lids and caps are positioned and tightened correctly.
- Sealing validation - ensuring that foil covers and anti-tamper seals are present and correctly applied.
Code validation
Machine vision solutions for code inspection are used to verify code presence, position, formation and readability, and sometimes to also provide code reading and matching. Such systems can automatically identify and reject containers or packages with missing, incorrect or unreadable codes to ensure only properly coded items are produced.
Codes that can be validated are date codes, batch codes, barcodes and 2D data matrix codes, all ensuring that non-compliance can be traced and acted on.
Label inspection
High-speed labelling of products, of all types, shapes and sizes, can result in a wide variety of possible defects. These defects can lead to labelling errors that can be harmful to a brand or even present liability issues for a brand owner. Labels can be inspected for label presence, wrinkles, tears, skewed labels, double labels, flagged or missing labels, as well as incorrect label pairs (back and front) on containers and packages.
Overwrap alignment is another form of label inspection in which wraparound labels are checked for straightness and proper position. With appropriate MV system design, a 360° inspection on round bottles can be performed.
Label validation
The importance of correct labelling on a food product grows each day as scientists and doctors discover more food properties that need to be identified and displayed: peanut allergies, gluten-free, salt content - the list is extensive.
Machine vision technology for label inspection can be set up to help ensure correct labelling, and packages and containers with incorrect or defective labelling can then be automatically rejected in the production line.
Incorrect product labelling on the supermarket shelves could result in costly litigation, or worse.
Packaging integrity
Inspection of recycled packaging equipment can be easily achieved using MV technologies.
In beer bottling lines it is common practice that the beer crates are checked for wholeness, and if the crate has any damages, such as a hole in the bottom, it can be rejected before a bottle is inserted and falls through.
A further application has been implemented by a leading confectionery company in Melbourne that uses MV to check that the confectionery trays are clean for re-use before reintroduction into the manufacturing process of the next batch.
Filling monitoring
Ensuring that the correct product is filled into the correct container, and that the correct amount is filled, can be achieved using MV together with the code validation technologies.
This application is used by a global pet-food producer with process plants in Australia. Here the importance of correct filling is even higher due to the fact that much of the production is exported to Asian markets where brand reputation is paramount. Product recall, in this instance, can be very costly, both in terms of product waste and brand damage.
Closure validation
Obviously the integrity of closures and seals on bottles and other containers is important for the quality of the product and the safety of the consumer. MV systems can be used to visually check the closures and seals for integrity.
A simple application, albeit in a related industry, is using MV technology to check that the correct colour cap is placed on the correct home brand bleach bottle at a Queensland-based contract bottle filling company. And, more importantly, that the sealing is perfect - leaking bleach in the boot of a car could have some unpleasant consequences.
Checking the closure’s colour and dimensions verifies that the right closure has been applied to the container, while visually checking liner formation and placement ensures the product is properly sealed and protected from contamination and leakage. In the same way, tamper seals can be checked to make sure they are not broken.
Sealing inspection
There are many other applications where the integrity of the sealing process is paramount. The sealing of all wet food needs to be checked - any small flaw in the seal will result in the product possibly leaking contents, and the definite likelihood of the product spoiling. Additionally, tamper-proof seals can be checked for correct application prior to packaging and a photographic record kept of every product inspected.
We have reviewed a small selection of applications for the use of MV technology. Further applications of machine vision in other markets include:
- Checking product quality - bruises, cuts, damage to fruit and vegetables, cracked eggs.
- Sorting fresh produce by size, colour and type - nuts, oranges, eggs, coffee.
- Checking for defective bottles and hygiene inspection in the bottle-washing process.
- Additional checks on bottle crates - logo inspection (correct crate for specified product), edge inspection to avoid sharp edges causing injury.
- Fish filleting using MV in conjunction with robots.
MV technology is developing as a rapid rate, providing faster, smarter and more accurate measurements. This evolution ensures that the food industry can apply the technology to more complex track and trace applications thereby assuring an even higher degree of quality assurance and food safety.
As the food and beverage authorities tighten the regulations, MV solutions are the best way to ensure compliance.
Advanced robotics in tomorrow's factory
Addressing the production challenges of complexity, customisation and openness.
Cracking the nut: robotic automation at Freedom Fresh
SCARA robots from Shibaura Machine have found a place in helping to package macadamia nuts.
Food plant expansion sustained by central robotic palletising system
A palletising system with eight robotic cells has been installed at Unilever's food factory...