Improvement and automation of cleaning processes in the  Pharmaceutical Industry

João Ludgero Henriques Pereira

Key information

Authors:

João Ludgero Henriques Pereira (João Ludgero Henriques Pereira)

Supervisors:

Tiago Faria; Maria Cristina De Carvalho Silva Fernandes (Maria Cristina De Carvalho Silva Fernandes)

Published in

January 27, 2021

Abstract

The cleaning process in-between production batches in multi-purpose pharmaceutical industries is an ongoing challenge since it represents downtime where value is not being produced to the customer. This thesis aimed to provide tools which promote the continuous improvement of the cleaning process of a specific spray-dryer unit, SD1. The Standard Work Methodology was applied, and the Cleaning Procedures were investigated to uncover the Top 3 Time-Consuming Operations and the Top 3 Sources of Variability. A Register Sheet was coupled with a Swimlane Map Continuous Improvement Tool to allow for the proper quantification of the cleaning operations and provide an automatic treatment of the collected data. A Suggestions Sheet was created to allow the operators to give feedback regarding the inaccuracies in the Cleaning Procedures and propose suggestions for improvement. A Continuous Improvement Loop was developed based on the two tools provided, the Register Sheet and the Suggestions Sheet, to tackle both of the identified problems: the impossibility of quantification of the cleaning operations and the lack of opportunity for the operators to give feedback regarding the Cleaning Procedures. The study on the potentiality of automated cleaning processes revealed that a decrease in the lead time of 36% could be achieved due to the parallelization of operations and the use of an extra CIP Tank. The Master Batch Recipe also promotes variability reduction since the control system performance is not relying on the operator’s proficiency with DeltaV which leads to a positive impact on predictability.