Quality Management and Real-Time Process Control.
The Sherlock Food Analyser monitors production lines with 100 times more accuracy than conventional methods.
Due to its modular design, the Sherlock Food Analyser can be placed anywhere in the processing line. From raw product receiving, peeler control, upstream and downstream of the dryer, upstream and downstream of the fryer to the finishing line.
Thanks to Chemical Imaging Technology (CIT® Gen.3), the Sherlock Food Analyser provides chemical data of the product being processed. In addition, high-resolution color cameras provide information on the optical properties of the products.
The InlineFOODLab 4.0 technology makes this data available to the quality management and process control system so that adjustments and optimizations can be made in the process line fully automatically in real time.
Sherlock Food Analyser is available in stainless steel, hygienic design and fully wash down cleanable. The space requirement is small and the integration into existing lines is simple.
Example: Dry Matter Content
The precise analysis of the product flow allows for highly accurate determination of dry matter content, leading to higher yields. Reducing the dry matter in the final product by 0.25% to 0.5% at a line capacity of 20 tons per hour can generate up to €1.7 million in additional profit per year.
Reliability
Unlike traditional methods, which take a few samples and provide results after up to 60 minutes, the Sherlock Food Analyser collects hundreds of thousands of measurement points per hour, offering the most reliable picture of the entire production.
Sampling
To monitor continuous product flow, only a few samples are usually taken and analyzed in the laboratory. These samples poorly represent the true distribution and mean value. Additionally, the time delay in receiving results means production errors are detected late.
Uncertainty
Empirical data show a 2.4% standard deviation in dry matter value. Due to high natural fluctuation, only a large number of measurements can reduce uncertainty, making the small number of laboratory samples insufficient.
Student’s t-distribution
The method to reduce the measurement uncertainty is scientifically proven by the student’s t-distribution under the null hypothesis.
Non-destructive
Unlike complex destructive sampling, the Sherlock Food Analyser measures the product flow continuously and nondestructively, recording over 100,000 values per hour. This allows real-time analysis of production, providing mean values and deviations.
This intelligent function is already available to you today on every Sherlock sorter.
It is the first step for the smart factory of the future.
Maximum Efficiency and Precision –
Discover the Benefits of Sherlock Food Analyser
- Significantly improved overview of the existing quality of raw and processed food
- Optimized use of raw materials and thus reduction of the raw material costs
- Optimized process reducing energy consumption, maintenance and labor cost as well as other production related costs
- Reduction of internal product holds and quality related complaints, one step closer to “zero waste production”
-
Permanent monitoring of the product relevant real-time parameters
1. Defect-Group Chart
2. Rel-Defect-Area Chart
3. Dry-Matter Chart
Some Application and Defect Examples
Technology that Makes the Difference
CIT’s chemical inspection capabilities brings us to new levels and possibilities of product inspection compared to any other optical eye out there such as lasers, various amount of cameras or Xray. Where other optical eyes focus on trading off the amount of bad in good vs good in bad; CIT’s approach is seeing and removing it with the highest reliability of the industry, no matter what product or defect and this at consistent levels even if there are seasonal product changes or product variety changeovers. CIT does not need any operator to constantly be monitoring and keeping the machine in balance.
For the first time, artificial intelligence in the most versatile form of Deep Neuronal Networks is applied in real time to analyse the collected image data. It creates algorithms that enable the detection of even the smallest visible product defects as well as defects invisible to the human eye, fast enough to eject them from high-speed product flows. This new disruptive technology opens unimaginable applications and performance levels for food processors.
InlineFOODLab 4.0 allows processors to optionally get quantitative chemical data of the product and the most reliable real-time inline quality data available in the industry. For example dry matter values in potato products. These chemical data can be combined with color or shape defect levels as well as with size values of the objects. Additionally, any foreign material can be recorded. This allows quality managers to have better tools to control the raw material as well as the finished product.
