I am exploring using Bayesian Networks to identify the best parameters within a system design, to improve its performance. I'm trying to find any case studies where this has been used successfully or unsuccessfully.

As an example consider an automatic assisted anti-collision system on a car. It could use some combination of ultrasonic, radar & IR seekers to sense and measure the closing speed with the car in front. These all have different costs / reliabilities and perform differently in different weather conditions. I am interested in understanding how effective Bayesian Networks can be in evaluating the trade space and identifying the best solutions to these types of problems.


2 Answers 2


As I can see you're searching for a case study. Learning from data is useful because you can learn both the conditional probability distributions (CPDs) and the structure of the network. The best resource I've read is a freely-available Microsoft tech report.

Have a look at the paper, maybe it can help you upto some extend. I'll suggest you to start with Chapter 7 of it.

Hope it helps. Cheers!


There is a paper that was published at Systems&Cybernetics journal.

See how they use Bayesian Networks with sensor fusion:

Active and dynamic information fusion for multi-sensor systems with dynamic bayesian networks.

Hope it helps.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.