Getting Data Smart
Sachin Bhandari
Driving IT Compliance Digitization, CSV, CSA, QA-IT Leader. Member, GAMP Global Steering committee.
What do you see in the image above ?
One has volumes of paper which has a printed code. The other has volumes of data which has been stored ~ about 5 Petabytes of data, that's a lot. Half a ton of hard drives.
We all believe that we need this data and current scientific situation is such that we would actually need this data. But then we believed that we needed those papers 50 years ago.
We in the industry specially Pharmaceuticals, produce a lot of data, tons and millions of it and the regulatory bodies ‘seemingly’ need this data to establish that systems and processes are reliable, have behaved the way they should and there is an evidence to it. This data needs to be saved and preserved for almost a decade in some cases.
The data which needs to be captured or saved is Raw Data, Process data, Batch Data, Validation Data and training Data, QA and QC check data and what now. With industry moving to automation and going paperless, is it really a smart move? IMHO we are producing more data it’s only that we are not printing it all the time so it’s not visible. It’s stored in compact hard drives and tapes.
It’s time that we start getting smarter on the data we produce and store. We need to find ways where the systems are so reliable once the sanctity is established (System Validated) and security controls proved, we get rid of
a. Audit trails
b. Access logs
c. Process logs
d. Change logs
e. Historical data
This data is huge and not only increases the cost and time consumed but requires significant energies to keep it in shape.
Current regulatory framework is such that we need the raw data and processed data to comply and then audit logs, access logs to support that your system and processes were not tampered with.
But then why can we redesign the validation and security framework/processes to make sure system works as intended, un-tamperable and is secured all the time.
It’s time to wear our thinking hats , and do some out of box thinking to be Data Smart.
IT (GxP) System Manager|||Commissioning & Decommissioning IT System, IT Validation activities, NuGenesis SDMS System,Electronic Logbook -Tulip, Veeva
5yHi Sachin, interesting and nice article😊,yes I agree with Neeru..DA & AI..
CSV_Validation Manager at Accenture
5yHi sir , Its nice article ... The Paper vs. Digital debate is often portrayed as the battle of Old vs. New. The pros of digitization are Miniaturization , ease of use, Labor savings, Easily traceable, portability , and version tracking... however cons are software compromised, hacked or otherwise destroyed by any other way.. The, paper records are always stored on site, and many find that they are easier to protect than digital files. And many people find it easier to alter / replace records when they are on paper .The space required to store all of your paper records as you grow can become overwhelming. It is always good to move towards digitized and make it completely secured ..
Founder at Tech Qualitas | Helping organizations thrive by embedding quality in every step.
5yHi Sachin, interesting article, good topic to debate and discuss. Data Analytics and AI is the answer to manage big data, few pharma companies using this to reduce processing flaws, improve production quality, increase efficiency, and save time and money. Using data analytics, compliance group are notified if any manufacturing sites goes below defined quality parameters wherein audit trail, access logs, change logs managed but not removed as I understand . This requires collaboration among vendor, process SMEs, IT and definitely initial investment. I identified this while working on requirement from one customer to get all sites compliant using data analytics.