Case Study: Genentech cuts algorithm development time from months to weeks with MathWorks (MATLAB and OPC Toolbox)

Genentech Uses MATLAB and OPC Toolbox to Build a Supervisory Control Algorithm Development Platform for Bioreactors

Genentech faced slow, error-prone development of supervisory control algorithms after moving to a more reliable distributed control system (DCS), which made reimplementing and debugging algorithms take months and sometimes caused costly fermentation failures. To address this, Genentech partnered with MathWorks (MathWorks Consulting Services) and leveraged MATLAB and OPC Toolbox to regain rapid algorithm development, simulation, and deployment capabilities.

MathWorks developed and deployed the MATLAB-based ECALS platform — an object-oriented control engine that communicates with the DCS via OPC Toolbox, includes a GUI, a bioprocess simulation model, logging, version control, and automatic-restart features — and integrated it into Genentech’s pilot plant (10–1000 L). The MathWorks solution cut algorithm development from months to weeks (one algorithm reimplemented in two weeks versus six months), enabled identification of potential errors in minutes, has run reliably for over a year, and has controlled more than 800 successful fermentations.

Genentech

Ryan Hamilton

Process Researcher