MIGRAPASS™ is the online version of the well-established software MIGRATEST Lite 2001 developed by Fabes Research Company from Munich , Germany . The reliability of this software developed in accordance with the legislation requirement and its user friendliness allows it to be recommended as the “ specifically tailored and user friendly” Migration Modelling software in the Practical Guide of the European Directives.

Here are abstracts of the Practical Guide for FOOD CONTACT MATERIALS :
A PRACTICAL GUIDE FOR USERS OF EUROPEAN DIRECTIVES .
(Updated to 15 April 2003)
File: Practical Guide SANCO D3/LR D (04.2003)  

ANNEX I
MATHEMATICAL MODELS 

1. Directive 2002/72/EC, Article 5, paragraph 4, provides the following clause:

The verification of compliance with the specific migration limits provided for in paragraph 1 may be ensured by the determination of the quantity of a substance in the finished material or article provided that a relationship between that quantity and the value of the specific migration of the substance has been established either by an adequate experimentation or by the application of generally recognised diffusion models based on scientific evidence . To demonstrate the non-compliance of a material or article, confirmation of the estimated migration value by experimental testing is obligatory.

2. A research project subventioned by the Commission (DG Research) has established the mathematical equations to be applied and the conditions for their application. The appendix 1 describes the result of this research.

3. Commission guidelines
The Commission recommends to the enforcement authorities the application, when appropriate, of the “modelling”, as a tool to avoid very long and expensive analysis.
However it reminds also that if there is discrepancy between the results obtained by experimental testing and by application of a generally recognised diffusion model, the judgement for compliance should be based on the experimental results.

Nota bene
Even if there is only a very low probability that the results of the experimental testing exceed the value obtained by the application of the “modelling”, any person that noted this discrepancy are invited to inform the Commission.

APPENDIX 1 TO ANNEX I
ESTIMATION OF MIGRATION BY GENERALLY
RECOGNISED DIFFUSION MODELS IN SUPPORT OF
EU DIRECTIVE 2002/72/EC

(« Migration modelling ») 

This document contains the conclusions of the European project SMT-CT98-7513
‘Evaluation of Migration Models in Support of Directive 90/128/EEC'
(now 2002/72/EC)

[…]

Procedures, practical applications and examples

For migration modelling as described above corresponding home made computer programs may be established and applied. For conveniency reasons, it should be noted that a specifically tailored and user friendly computer program is available on the market (MIGRATEST Lite 2001, FABES GmbH Munich , Germany ) […]

Two principal procedures can be followed with the model:

As a general rule: In cases where the migration estimation scheme outlined above leads to results which are above the legal limits (SML), an experimental test of compliance is compulsory.

[…]

Optimising compliance control
Migration models can be used to optimise compliance control strategies. In fact there is an infinite variety of packages to be tested, with different geometry, size, type and thickness of polymer, food, shelf life of food, temperature etc and the effort to check all these influences is huge. By using the migration model one can find out relatively easily the worst case system, (covering all the less severe situations) and then reduce the experimental migration measurement just to those cases or samples where it is possible to be in non-compliance at all.
[…]
Having the experimental result from one selected test, all the different conditions and parameters, as mentioned above, are then modelled without a supplementary experimental test. In this manner a significant increase of the quality assurance is possible with many samples controlled during the same time compared to one experiment without the additional tool of modelling. Of course, the nature of the plastic material must be known as well as the initial concentration of the concerned migrant (known or disclosed by producer or experimentally tested). If not, a reasonable upper limit can be used.