Swiss auditors find that COMCO's bid-rigging screen works at low cost

The Swiss Federal Audit Office found that COMCO's machine-learning screen for bid rigging in construction gave concrete, encouraging results in real cases. COMCO spends under CHF 10,000 a year on it.

  • Not rated an organization

    The audit office found encouraging results in concrete cases but could not quantify the added value of a tool used about once a year.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent audit report supports the benefit.

The system’s part
Not isolated. The audit office gives no comparison of detection with and without the machine-learning methods. The screen combines statistical tests with logistic regression, random forests and neural networks, and specialists interpret its results.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
COMCO bid-rigging screen for construction tenders (statistical screens combined with machine learning, run as a script)
Made by
Secretariat of the Swiss Competition Commission (COMCO/WEKO), with the University of Fribourg from 2019
Used by
Swiss Competition Commission (COMCO/WEKO)
Where
Switzerland
Stage
In a pilot
Date
, when the source was published
Kind
Program
Broad benefit
Analyzing Data or Information, Evaluating Information to Determine Compliance with Standards
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Audit report, by authors with no tie to the system.

    Contrôle fédéral des finances / Eidgenössische Finanzkontrolle (CDF/EFK). Audit de l'utilisation d'une application avec intelligence artificielle: Commission de la concurrence (COMCO). CDF-25407, version with responses dated 11 February 2026, published 22 April 2026. Section 2.1, p. 11.

    La solution a été testée par la COMCO et par les pairs, et utilisée dans des cas concrets. Les premiers résultats de l’utilisation de l’outil intégrant l’IA montre des résultats concrets et encourageants quant à l’utilisation de la technologie de l’IA.

Details

AI system

Workflow

COMCO combined classic statistical screens with machine learning on data from construction tenders. Specialists who can interpret the results ran the script and reviewed its output.

Follow-up

Conditions for replication

Cantons and communes must supply tender data, which they do mostly on a voluntary basis. A person decides whether to open an investigation.

Entry details

Sector

Competition enforcement / public procurement.

Purpose

The Secretariat of the Swiss Competition Commission built the screen, with help from the University of Fribourg, to detect bid-rigging cartels in construction tenders.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Swiss auditors find that COMCO's bid-rigging screen works at low cost (AIBD-0714). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0714.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions