data analysis

Analyzing the mountains of data your business generates can mean the difference between "just getting by" and success. It's also the best way to get a deeper understanding of your business. 



Being a student of your data is perhaps most important in manufacturing. Sometimes traditional techniques like regression and ANOVA are the right tools for the job. Other times we're looking for complex interactions between equipment. Sometimes we just want to answer questions like:

  • We went 7 days without a loss on a certain process step. How often should we expect to see that?
  • How can we tell if the process change we did improved our long-term yield?
  • What percent of the time should we expect a yield above X%?

In the grueling world of manufacturing, sometimes it's useful to have a fresh set of eyes looking at a a problem. Experienced eyes are even better. 

Related Content: Using Bayesian Analysis to Predict Process Yield

Monte carlo simulations

Monte Carlo simulations are a powerful way to predict the effect of a change in a complex process. The applications are wide ranging from medicaid to manufacturing to sports analysis.

Monte Carlo is especially well suited for healthcare due to the complexity of the calculations and the uncertainty of the caseload over time. Read more here.

Related Content: What's an ADL Point Worth?

Related Content: What's A CMI Point Worth?

Kentucky's pre-tournament odds of winning the 2016 SEC Tournament: 1.5:1

Kentucky's pre-tournament odds of winning the 2016 SEC Tournament: 1.5:1


Bayesian analysis

When appropriate, I prefer to use Bayesian analysis versus traditional methods. The results are more robust because we don't need to depend on things like normal distributions and we can take prior knowledge into account. This allows us to handle an outlier without ruining the model.

It's not a panacea however. Let's discuss your project and figure out what best suits your needs!