(146e) Exceptional Event Managemnt Applied to Continuous Pharmaceutical Manufacturing | AIChE

(146e) Exceptional Event Managemnt Applied to Continuous Pharmaceutical Manufacturing


Venkatasubramanian, V., Purdue University

One of the important challenges in effective real time process management

is the implementation of intelligent systems that can assist human operators

in making supervisory control decisions, instead of simply sounding an alarm

when process variables go out of range. Operator failures to exercise the appropriate

mitigation actions often have an adverse effect on product quality,

process safety, occupational health and environmental impact. The economic

effect of such exceptional events is immense; an estimated $20billion/year losses

in petrochemical industry have been reported. The challenges and opportunities

for improvements are even larger in the pharmaceutical manufacturing domain

because so much of the processing involves particulate and granular systems

whose characteristics tends to be more problematic than that of fluids.

The development and manufacturing of pharmaceutical products are governed

by strict safety regulations but with the advent of Process Analytical

Technology (PAT) initiative advanced by the FDA; gives the pharmaceutical

industry an opportunity to apply various systems engineering tools. Early detection

and diagnosis of process faults while the plant is still operating in a controllable

region can help avoid abnormal event progression, production disruptions

and productivity losses. An EEM framework has been developed to deal with

fault detection, diagnosis and mitigation of conditions that result from process

anomalies. The framework developed uses a combination of Principal Component

Analysis (PCA), Signed Digraphs (SDG) and Qualitative Trend Analysis

(QTA) in applications involving continuous pharmaceutical product manufacturing

line. An ontological database has been created to maintain records of the

signatures of a number of exceptional events typical of a dry granulation line

and their associated mitigation strategies. In trials on pilot scale equipment we

have found that the EEM framework was able to detect and diagnose several

types of faults within a few seconds of their inception and to provide mitigation

advisories to the operator.