... in which the catalyst was prepared by 1:1 feeding ratio of Au:Pt and was
thermally treated 250 °C in air for 30 min, which showed a particle size of 6.8 ±
1.4 nm. The data for that AuPt/C (40% wt) catalyst obtained under the condition of
a much lower dischargingcharging current density (0.04 mA/cm2electrode)
showed a discharge voltage of ∼2.7 V and a charge voltage of 3.6 V on average,
and a capacity value of 1200 mAh/gcarbon. For the Au49Pt51/C catalyst (as
shown in Figure ...
Prediction of biopharmaceutical facility fit issues using decision tree analysis
The impact of upstream and downstream process fluctuations in bioprocesses becomes more pronounced during technology transfer of pilot scale processes into large-scale facilities. Mismatches in equipment sizes upon scale-up combined with process fluctuations can lead to discarding expensive product. Understanding and predicting the root causes of product loss is critical to determining the necessary facility modifications required to meet manufacturing targets. In this work, a decision tree classification method, CART (classification and regression tree), was used to interrogate stochastic Monte Carlo simulation datasets that mimic batch fluctuations in industrial biopharmaceutical batch processes. The resulting binary decision tree not only reveals the impact of process fluctuations on product mass loss but also extracts a series of if-then rules on the critical combinations of factors that lead to unacceptable mass loss levels. The decision tree analysis therefore allows rapid prediction of the likelihood of facility fit issues as well as the batch characteristics that result in product loss.
Prediction of biopharmaceutical facility fit issues using decision tree analysis
Yang Yang, Suzanne S. Farid, Nina F. ... Yang Yanga, b, Suzanne S. Farida and
Nina F. Thornhillb, aEPSRC Centre for Innovative Manufacturing in Emergent ...
