The Alexandra Institute // Cases // COBE – Confidential Benchmarking

COBE – Confidential Benchmarking

Case

COBE – Confidential Benchmarking

Compare confidential information while keeping it secret!

Businesses can perform better benchmarking with new, secure solutions based on state-of-the-art cryptography. 

In the COBE project (Confidential Benchmarking) economists and computer scientists have jointly developed prototypes that perform benchmarking based on Secure Multiparty Computation (SMC) and economic theory. The prototypes were tested in four specific applications.  

Benchmarking is widely used by industry to improve the performance of organisations, systems, work processes etc. on many levels of the economy. Benchmarking relies on the willingness of the involved parties to supply sensitive data to a third party that performs the benchmarking process. Benchmarking may therefore be a costly activity that requires strong security procedures. 

However, modern cryptography holds promises for a completely different and better solution. The COBE project has therefore developed a number of prototypes that perform benchmarking based on Secure Multiparty Computation and tested these in practice.

Modern cryptography makes it possible to develop benchmarking solutions that are both cheaper and safer than traditional solutions.

The objective of the COBE project has been to demonstrate that Secure Multiparty Computation can be used to perform much more complex computations than those performed in previous research. The project developed new cryptographic protocols and algorithms as well as a number of prototypes.

A demo has been made available to the public to enable potential users and others interested to see how the system works.

 

Secure Multiparty Computation (SMC) is a technology that allows you to compute on encrypted numbers.

SMC allows a number of servers to jointly compute any function without learning the inputs to the function. This is achieved by providing each server with only partial information on each input. The partial information given to a single server (in some cases even several servers) reveals nothing about the actual input. Nevertheless, using the right cryptographic tools, the servers can work together and compute the result.

With SMC, all input data are supplied in encrypted form and remain encrypted during processing. The results produced are encrypted as well, which means that the only result that is made public is the one that the parties agree to have decrypted.

Thus, SMC fundamentally parts with the idea that one needs to trust in specific individuals or organisations to benchmark private information.

 

Extensive knowledge transfer between industry and academia has taken place throughout the project.

A number of knowledge dissemination activities have been carried out to make Danish companies aware how they can apply Confidential Benchmarking to improve their business. This has been facilitated through an annual seminar and via the Alexandra Institute’s extensive network.

In collaboration with a number of industry partners, the COBE project tested the prototypes developed during the project in these application scenarios:

  • Evaluating environmental impact in shipping 
  • Increased transparency in the interbank market 
  • Bonuses based on relative performance 
  • Eliciting unknown unit costs in multi-output production 

The applications listed above are only examples in a much wider range of scenarios where Secure Multiparty Computation-based solutions create value in society at large.

We have developed solutions for the agricultural and banking sectors.

We can help your business too.

Contact
IT-City of Katrinebjerg

Åbogade 34

8200 Aarhus N
+45 70 27 70 12
.