The Data Centric Architecture of a Factory Digital Twin
5.2.5
RESULTS
Figure 5-8 illustrates the solution to the base case problem , with no random delays on the chromatography columns . Following the process from front end to back end ( top to bottom ), we see the three parallel flasks that feed the three parallel Bioreactor trains . The initial flask tasks are staggered due to a resource constraint that prevents multiple Inoculation tasks from starting at the same time . As the Bioreactor train starts up , the 10,000-liter bioreactor quickly establishes the rate of the whole train . Material awaiting processing is held in the two smaller bioreactors . Effluent from the 10,000-liter bioreactors is held in the Fuge Feed Tanks until it is processed in the centrifuge .
We can see the three sequential chromatography columns processing with the yellow repack tasks interspersed at the frequency required for each of the columns . The last stage of the process is the relatively fast Filter task at the bottom of the chart .
The 10,000-liter bioreactor is the rate limiting step as we can see by the fact that there are no gaps on this reactor . It is also evident that the Revo column is very nearly rate limiting , only small gaps exist . The pink storage tasks reveal a wealth of information about the nature of this process .
Figure 5-8 : The base case solution .
116 February 2025