From food to pharmaceuticals, Borde adapts to the specific needs of each industry by harnessing the power of machine learning to continuously improve and deliver unprecedented results.
Explore the operational possibilities, that have until now been out of reach!
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.
Can Borde recognize a wide range of product defects and foreign objects?
The Borde Inline System combines patented AI software (aka Borde Inline Platform) and AI hardware to recognize a wide range of product defects and foreign objects. Borde Inline Advisor is our premier AI inspection software, that is designed from ground up in the Silicon Valley- not only tohandle 1 Million objects per second so that each foreign object and defect canbe inspected, but also provide the most accurate inspection system to avoidrecalls and reduce human labor.
Is the Borde Inline System a single sided inspection?
Yes, the Borde Inline System conducts single-sidedinspection of the product that flows past the system. However, as Borde Inline sees every item, in direct comparison with sample data from customers, Borde Inline achieves high accuracy across defect classes with the added bonus of seeing every piece of foreign material and outperforming manual sampling fordefect classes that are less likely to show up in a manual 500 gram sample (gummy, shrivel, etc.). Because it is single sided on all items, Borde statistically normalizes some defect categories and those results have been proven to matchon exact bin count or bigger sample size QC measurement.
What volume can the Borde Inline System support?
The Borde Inline System is volume agnostic, capable of inspecting up to 20 tons an hour per line. Processors should feel comfortable running their lines as fast as they want with no data/detection loss from processing speed.
What defects is the Borde Inline System capable of detecting?
The Borde Inline System maps to the governing standard bywhich the product is graded. As such, each model contains data for whatever product it is inspecting. In almonds as an example, the Borde Inline System captures and categorizes kernels into edible kernels and multiple dozens ofdefect and foreign material subcategories. It then returns that data with aUSDA grade and breakdown by piece count and percentage by weight.
Can the Borde Inline System size product?
Yes, the Borde Inline System can size product. One of ourmost popular deployments for almond processors is on the sizer, returning bin level quality control reports with accurate CPOs through use of a proprietary sizing model. For walnuts, the Borde Inline System can also size product intohalves and pieces with a high degree of accuracy.
Does the Borde Inline System require human interaction?
One of the benefits of the Borde Inline System is itsability to be fully automated. Integration with check weigh scales, PLC signals upstream or downstream and 3rd party ERP/inventory management systems creates a seamless end-to-end process flow, capturing images and creating quality control reports on a lot, bin or box level without the need for a human in the loop.For lines where PLC automation is not available, a manual start/stop of the Borde Inline System is available through the UI that all Borde customers have accessto. Database export remains automated at the end of the report.
Can the Borde Inline System integrate with my sorting machines?
At present, the Borde Inline System operates independently from sorting machines. However, the Borde Inline System captures and returns quality control data in real time anywhere in the plant, allowing for faster operator action in adjusting sensitivities on sorting systems to account fordeviations in product quality. Looking forward, Borde expects the market tomove towards allowing real-time closed loop feedback to sorting systems fortighter control and automation within the plant, minimizing the need for expensive human labor.
How does Borde Inline compare against manual sampling or other legacy 500 gram sampling AI systems?
The Borde Inline System tends to outperform manual sampling or other 500 gram sampling AI quality control systems. As both operate onoutdated sampling methodology of 500 grams on 2200lbs, the quantity of product inspected is much more representative of the product that is actually runningthrough the plant. Frequently, foreign material and certain defect categories will not show up at all in a traditional sample, but that doesn’t mean theyaren’t there. When they do show up in a sample, extrapolated out, the numbers indicate that there are pounds of foreign material or certain defects in a binwhich is also not true. The Borde Inline System’s ability to inspect every kernel ensures that processors are receiving highly accurate quality controldata that is significantly more representative of the overall product quality. What we find is that as the sample size or sample rate increases, the average numbers from the increased sample size or rate trend more closely to thenumbers the Borde Inline System returns. Additionally, the Borde Inline Systemis a fully in-line solution that requires minimal to no change to existing process flows. The Borde Inline System can be flexibly deployed anywhere in the process flow where a single layer flow is present or where processors need quality control data, capturing and returning quality control data in realtime. Compare that to existing sampling methodology or other AI technology that is not in-line and still requires a human to pull a sample from the line.
How long does a Borde Inline system take to install?
A single Borde Inline System can be installed in as little as 6 hours. For larger deployments such as a full sizing line, the installation takes about a week.
What can the Borde Inline System detect?
The Borde Inline System can detect any defect or foreign material the human eye can see. Our models are trained to detect all relevant defects and foreign material particular to the crop. The Borde Inspect System can also grade based on color.