Customer-based prediction method and system using near infrared reflectance spectra of materials

• US 6,907,351 B2
• Filed: 08/01/2001
• Issued: 06/14/2005
• Est. Priority Date: 08/01/2001
• Status: Expired due to Fees

• Current Assignee: Adisseo France S.A.S. (China National Chemical Corporation)
• Original Assignee: Aventis Animal Nutrition SA (China National Chemical Corporation)
• Inventors: Bodin, Jean-Christophe, Julia, Thierry
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Barbee, Manuel L.
• Application Number:
  US09/918,512
• Publication Number:
  US 20030028328A1
• Time in Patent Office: 1,413 Days
• Field of Search: 702 22- 25, 702/28, 702 30- 32, 702/108, 702121-123, 702/127, 250/339.09, 250/339.12, 436/8, 436/86, 436/164, 436/171, 705/1, 705/10, 705/16, 705/20
• US Class Current: 702/23
• CPC Class Codes: G01N 21/3563 for analysing solids; Preparation of samples therefor
• CPC Class Codes: G01N 21/359 using near infrared light
• CPC Class Codes: Y10T 436/10 Composition for standardization, calibration, simulation, stabilization, preparation or preservation; processes of use in preparation for chemical testing

Abstract

A method and system for predicting the content level of components in materials based on the response of the materials to near infrared radiation. One embodiment comprises electronically receiving a near infrared reflectance spectrum of a material from a customer, predicting the content level of a component in the material based on the spectrum, and electronically reporting the prediction to the customer. Another embodiment includes the exchange of the customer request and prediction report on a Web site or by electronic mail. Other embodiments include the prediction reports, and uses of the predictions, for example in quality control and toxicity evaluation.

Claims

1. A method comprising:

electronically receiving a request from a customer to predict the content level of at least one component in a feedstuff, wherein the request includes a near infrared reflectance spectrum of the feedstuff and the component is selected from proteins, total or digestible amino acids, gross or metabolizable energy, total or retained phosphorous, an impurity and a toxin;

comparing the spectrum to a specific database calibration that correlates known content levels of the component in other feedstuff to known near infrared reflectance spectra specific of the other feedstuff

predicting the content level of the component; and

electronically reporting the prediction to the customer;

said method further comprising determining that an outliner exists when the prediction of the content level of the component has a degree of error that exceeds a predetermined threshold value; and

providing the prediction report to the customer at no cost when the prediction of the content level is determination to be an outlier.

2. A method as claimed in claim 1, wherein the calibration includes at least one correlation between a known content level of the component determined by in vivo measurement and a near infrared reflectance spectra of the feedstuff containing that component.

3. A method as claimed in claim 1, wherein the feedstuff is cereal, corn, soybean cake, oleoproteinaceous flour, animal meal, animal byproduct, fish meal, cereal byproduct, or silage corn.

4. A method as claimed in claim 1, wherein the at least one component is total or digestible methionine, lysine, cystine, threonine, tryptophane, valine, isoleucine, phenylalanine, histidine or arginine.

5. A method as claimed in claim 1, which comprises presenting one or more menu options for selection by the customer in making the customer request.

6. A method as claimed in claim 5, which comprises presenting to the customer menu options for the report format of the prediction report.

7. A method as claimed in claim 5, which comprises presenting to the customer menu options for the category of feedstuff represented by the spectrum.

8. A method as claimed in claim 5, which comprises presenting to the customer menu options for one or more components whose content level is to be predicted.

9. A method as claimed in claim 1, which comprises reporting the prediction to the customer within 24 hours of the customer request.

10. A method as claimed in claim 1, which comprises reporting the prediction to the customer within 10 minutes of the customer request.

11. A method as claimed in claim 1, wherein the customer request and the prediction report may be exchanged 24 hours a day.

12. A method as claimed in claim 1, which comprises providing the customer with one or more identity and/or security codes for use by the customer in making a request.

13. A method as claimed in claim 1, which comprises verifying the one or more identity and/or security codes upon receipt of a request.

14. A method as claimed in claim 1, which comprises storing the customer request and prediction report of one or more customer requests.

15. A method as claimed in claim 14, wherein the stored information may be retrieved by the customer upon request.

16. A method as claimed in claim 15, wherein the stored information may be retrieved by the customer on a Web site.

17. A method as claimed in claim 1, further comprising providing a central database with calibrations based on samples taken from geographically diverse regions, wherein the calibrations of the central database comprise the specific database calibration for predicting the content level of the component.

18. A method as claimed in claim 1, further comprising providing a central database with calibrations based on materials produced in different seasons of the year, wherein the calibrations of the central database comprise the specific database calibration for predicting the content level of the component.

19. A method as claimed in claim 1, further comprising:

presenting, when the prediction of the content level is determined to be an outliner, an offer to the customer to perform a measurement of the content level of the component based on a sample of the feedstuff; and

in response to an acceptance of the offer by the customer, measuring the content level of the component in the sample of the feedstuff and reporting the results of the measurement to the customer.

20. A method as claimed in claim 1, further comprising correlating a near infrared reflectance spectrum of the sample of the feedstuff with the measured content level of component to produce calibration data and adding the calibration data to a central database to enrich the database for handling future customer requests.