Tastel and Tastel+

previous summary next

b - Development of a product profile

The profile techniques are very different in sensory analysis and can be used to provide many answers, one can cite some of them:

  • Assistance in product development with identity cards sensory products according to their different components in appearance, smell or taste.
  • Aid for the choices of materials or processes, by defining the sensory characteristics required.
  • Follow up of divergences of quality characteristics in relation to 'standard', follow-aging ...

We find the different phases of conducting a test, easily implemented in Tastel with basically three steps, questionnaire preparation, data collect, and finally calculations with the possibility of requesting an automatic report generation to help in formatting and interpreting the test results.


1. Questionnaire preparation

It is always possible to take back a questionnaire already used, but we will do a complete test using the example of a test score of four wines, although it is possible to choose different protocols (complete, incomplete balanced randomized, Latin square, ...), we will request a complete design presentation.

First just very simply define the parameters of the questionnaire, they are, for example, the type of data collect: compared or monadic, the scale to use: structured or continuous, the automatic code generation, presentation samples, product names, names of the descriptors, or sensory attributes, the wording of possible suggested comments, and possibly other optional parameters.




It is of course possible to add screen presentations with images, sounds, videos, very simply by showing by example RTF file type.



2. Data Entry

The preparation completed, the system can edit the questionnaires directly in Word©, they allow a scan if is needed a data collection with a scanner or via Internet forms Word©. These questionnaires are modifiable and can be customized with additional explanations, pictures, ... while retaining their auto fill capacity.

The data collect is usually displayed for the judges as shown in the picture below (different ways of presentation are of course available).



3. Data processing

Once the entry is complete, the data are stored in a database, and just ask the treatment of these results, different parameters are required as sessions to deal with (1 to 50, or more than 200 in tracking profiles), comparison or not (adding) these sessions, thresholds of significance (calculated thresholds are also provided), the scale, standardization of data or not ... but also the product selection, judges and variables to process.


Options for handling missing data can also be applied if data are missing in these data.


Graphic results

Various statistical techniques can then be implemented and displayed in the results, such as star charts, comparisons in pairs.




All graphs present in TASTEL can be issued in EXCEL© format automatically, allowing great flexibility and customization of graphical output.


The first graphical results can be complemented with tests based on averages (Student Tests, Analysis variance), or with tests based on rankings (Willcoxon tests, Friedman tests), on whether the differences are significant or not.

An example of results of analysis of variance is given below:


The multi-dimensional analysis (CPA, CFA, discriminant analysis, STATIS, classifications, ...) can take into account all the descriptors for a synthetic image results, so you can see results from a Principal Component Analysis on the product details or with envelopes of significance.






Report Generation

All these results can be formatted and interpreted automatically by requesting a report generation, a report consists of a merger between a predefined and user text which can be changed and results from calculations. The user specifies the results he wants, knowing that this application may be a model that can be applied to any further processing:

The results are as follows always on our 4 wines ; out of these results in report generation can take place directly after the data entry, and does little more than a minute.







One can observe that the system automatically detects the type of statistic (parametric or non-parametric) to implement for the comparison of product synthesis, it is also possible the system automatically shows an interpretation of the results of the PCA.

All these results can be automatically formatted in Excel©, and therefore can be picked up for more personnalized communications and can be for treated in different media.



Top