RBMV Sample selection

Province of Limburg

This algorithm automatically selects cost lines from partner declarations in Interreg Meuse-Rhine projects for sample verification. The sample selection is based on the principle of risk-based management verification (RBMV). The algorithm combines selection of the highest amounts with monetary unit sampling (MUS) to include both risk-based and random cost lines in the sample.

Last change on 13th of November 2025, at 14:35 (CET) | Publication Standard 1.0

• Publication category: Other algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

Organisation and business operations

Begin date

05-2025

Contact information

[e-mail]

Link to publication website

https://www.limburg.nl/bestuur/open-overheid/algoritmeregister

Responsible use

Goal and impact

The purpose of the algorithm is to efficiently and objectively sample cost lines to be checked within the Interreg Meuse-Rhine programme. Programme staff use the algorithm to check project partners-organisations whose costs can be selected for further checking, depending on the risk profile (manually determined) of the project partner. The organisations involved experience the impact of the algorithm as certain cost lines submitted by them can be selected for further checking.

Considerations

Automation of sample selection makes the process more efficient, transparent and reduces the risk of human error. As a result, sample selection is done faster and according to objective criteria. Disadvantages are dependence on the correct functioning of the algorithm; for this reason, extra attention has been paid to the implementation of the MUS principle and randomisation in Excel. The elaboration is in line with the RBMV approach adopted by the programme's managing authority.

Human intervention

The outcome of the sample selection is always checked and assessed by staff of the Interreg Meuse-Rhine programme. They can see the results and make enquiries or additional checks where necessary. Intervention and adjustment are possible if irregularities are found.

Risk management

The risks of automated selection were actively considered, especially the quality of randomisation and MUS implementation in Excel. Where necessary, results are additionally checked. If the RBMV approach changes, the algorithm is adapted. No personal data are processed.

Elaboration on impact assessments

No personal data and/or privacy-sensitive information will be used

Operations

Data

The algorithm uses as input cost lines and edge information from the JEMS system (Interreg Meuse-Rhine monitoring system), supplemented by the risk profile determined per partner (determined manually). No personal data or special personal data are processed.

Technical design

The cost lines and associated information are exported from JEMS to the first tab of an Excel template. On this tab, the user selects the partner's risk profile. Using PowerQuery, separate tabs are filled for personnel costs and other cost categories. For personnel costs, the highest cost lines are selected based on the risk profile; for other cost categories, both high cost lines and a number of random lines (based on MUS) are selected. The result is an orderly sample that is easy to control.