Russey, D., Saint Arnold Brewing Company, Houston, TX, USA
Technical Session 7: In-Process QA
Wednesday, June 07, 2017
8:15–10:00 a.m.
Everglades A
With the recent boom in craft beer, many craft breweries cannot afford the cost of automated QAQC software, without which tracking trends in big data sets can become complicated. This presentation will introduce R and associated free software as a QAQC tool for breweries. R, and Rstudio, are both open source software that are free to download and use with no monthly fee, no recurring costs, and no advertisements. R is one of the fastest growing programming and statistical languages, and is used across many industries because it is free and highly versatile. Learning a programming language can be daunting, but R’s open-source nature has resulted in an incredibly helpful community with numerous blogs, forums, and channels that answer questions both simple and complex. Here, applications for R in a brewery will be reviewed (e.g., month to date reports, fermentation tracking), and a template will be provided for R novices to quickly adapt current Excel files to perform some basic statistical analyses with minimal adjustments. The template presented was developed to facilitate QAQC efforts to optimize brewhouse efficiency, but the template and analyses described can be generalized to hop trials, mash temperature adjustments, and many other recipe and process changes. As a technical example, the data presented here illustrate how this software facilitated brewhouse optimization trials, in which we achieved a 5% grain reduction and faster lauter times simply by altering the allocation of liquor during mash in and sparging. The emphasis here, however, is how using R aided our ability to plan, execute, and successfully analyze and validate our adjustments, and how one simple script can be used for numerous types of experiments. Decision making can be a difficult process, especially when it involves altering brewing processes or tweaking successful recipes. The goal of this presentation is to provide brewers with a template to build a reusable, personalized script to simplify common brewhouse statistical analysis and provide confidence in decision making for process improvements without the need for extensive programming experience. The open-source nature of R would mesh well with the largely cooperative nature of the brewing industry, since complex statistical processes, functions, and packages in R can be shared and reproduced. Ultimately, beer quality could be collectively elevated by fostering more transparent, scientific approaches to beer QAQC.
Drew Russey received a Ph.D. degree in biology from the University of Houston in August 2014. His dissertation focused on adaptive evolution of functional, multivariate traits resulting from natural and artificial selection. He joined Saint Arnold Brewing Company as a laboratory technician in August 2014 and was promoted to laboratory manager. This role includes aiding the QA/QC team in their statistical analyses of raw materials, product trends, and various brewhouse projects.
View Presentation | 
Download accompanying code