Solutions in Action
Case Study: Date Sorting Using Automated Internal Quality Analysis
Challenge: Food processors and sorting machine manufacturers demand fast, accurate, robust and simple optical sensing systems to ensure consistent quality and improve process efficiency
Solution: With robust hardware, application insight and advanced data analytics from Ocean Optics, food processors and sorting machine manufacturers are improving the quality of fruit and other foods throughout the food chain
Traditionally, food sorting has been managed using visual inspection, although the introduction of spectroscopic scanning has added a more sophisticated level of analysis, with some instruments able to penetrate the fruit peel non-destructively to determine parameters including fat content, sugar values and moisture levels. Also, spectroscopy techniques can be integrated into the process stream to measure more samples, at a faster rate, than with manual inspection.
Case Study: Lugo Machinery & Innovation, a leading supplier of fresh produce sorting products, approached Ocean Optics seeking an alternative to their manual method of sorting date fruit by moisture level. Their goals: Automate the sorting process to eliminate all manual inspection, and perform the measurements rapidly and non-destructively. What’s more, the customer timeline was very short, with only four months to develop a solution in time for date season.
Feasibility tests were performed on date samples from Lugo, which quickly showed NIR correlations to moisture levels in the fruit, helping to determine the choice of system hardware. This setup was used onsite to analyze a much larger sample set, then used for training data to develop proprietary machine learning algorithms.
Because Lugo had little experience with spectroscopy, they focused primarily on identifying the typical moisture peak for dates. But based on our experience analyzing other fruits and vegetables, we extended the analytical range to include broader patterned spectral features, which would help us to develop machine learning algorithms. This approach — broadband versus discrete wavelength spectral analysis – yields more accurate results and makes the data less susceptible to deviations related to optical interference.
The date-sorting system was integrated into a conveyor-belt setup with algorithms running on a devoted PC. The data fed a programmable logic controller (PLC) that triggers a valve, sorting the dates according to moisture level. This beta platform was used to refine and optimize the algorithms before final implementation.
Today, the fully integrated system scans 5 dates per second and is entirely automated. This has reduced overhead costs, improved safety and allowed Lugo to focus on refining other aspects of the sorting process. Also, with advanced statistical models now established, Lugo will be able to develop additional analyses without the need for modeling expertise.
With the fusion of spectroscopy, statistical modelling and machine learning architecture, sorting machine integrators and food processors can build more efficient sorting and grading systems.