Solutions
Platform
Community
Innovation
Company

Author: editor firma

July 30, 2021

How AI can Help Defeat COVID – A Glimpse into a United Nations Webinar

AI/ML
While we are still in a global pandemic, the sun is around the corner and advanced technologies help accelerate that timeline.

After we published our blog in January, How Advanced Analytics and AI can Help Lead Along the Path of Global Population Vaccination, where we discussed the main challenges we need to overcome to scale manufacturing and supply chain enough to inoculate the world against COVID-19, I recently had the pleasure to present in a webinar to the United Nations Industrial Development Organization (UNIDO) regarding the role that technology can plus in COVID-19 recovery challenges and impacts on biopharma. We’ve already talked a little about the role that artificial intelligence can play in increasing efficiency and expediency while reducing risk. Months later, there has been tremendous progress but we now have a world where some have access to the vaccine and some do not. At Aizon, we believe we technology has a role to play to further accelerate this. Take a look at this short presentation I gave to the United Nations on this topic.

by Toni Manzano, Chief Science Officer and co-founder of Aizon

May 3, 2021

Digital Twins – Seeing Double with a Predictive Eye

AI/ML Continuous Manufacturing
There are a multitude of definitions and use cases for a digital twin in pharma and biotech manufacturing. In the simplest terms, a digital twin provides an in silico model of the physical asset or process. Digital twins are employed in product and process development to facilitate agile development, technical transfer, and process improvement while reducing waste, positively impacting quality, and bringing a product to market faster.

To make this concept more tangible, we can imagine a “smart” automobile, such as a Tesla, for example. While the car itself is continuously producing and transmitting data around energy consumption, performance, maintenance, and more, it is also capable of capturing the environmental and passenger context– driver behavior, road and weather conditions, driver preferences, etc. The car and the context are inextricably linked and the insights coupled from the outside in and inside out enable better management and ultimately, a better driving experience. This holistic capture of data, processes, things, and people (the so called, ‘Internet of Everything’) can be used to create a digital model that predicts and recommends outcomes based on changes within the car or context. We can think of this as a digital twin.  

A digital twin for a biopharma process works in the same way. Take for example a bioreactor unit that is either standalone, part of a multi-unit process train, in process development, or active in commercial production. To create a digital twin with the goal of accelerating process development and optimization, all data from the equipment itself– raw materials, utilities, external context, and more, are required for the model to be meaningful. All this information must be integrated in the cloud in order to enable a thorough contextualization and close the essential feedback loop that is required to empower the model with continuous updates to adapt to the new scenarios in real-time. 

This ‘feedback loop’ can only happen with powerful Artificial Intelligence (AI) due to the complexity of the data involved and the requirement to generate valuable and actionable insights. The result of this approach is a full and intelligent digital version of your bioreactor that is able to proactively monitor its operation, identify the relevant factors that impact the product quality, process safety, and throughput. And it doesn’t stop there. Because the digital twin knows “everything,” it is able to provide real-time predictions on every meaningful information and the impact cascaded by the changes in the process. The outcome is, actionable insights that enable proactive decisions to avoid deviations, reduce cost and material waste, improve performance and product quality, all while constantly maintaining safety and compliance.

As we are talking about pharma, GxP compliance throughout the lifecycle of the data, the AI models, and the digital twin application is absolutely critical. This generates complexities, especially when the actionable recommendations must be applied in real-time like in continuous manufacturing and Continued Process Verification (CPV).

With that in mind, selecting the appropriate technology and defining the best approach to develop your digital twin is vital for a successful implementation.
Digital Twin

Planning for your Digital Twin Strategy – 4 Best Practices

  • Ensure GxP-compliant digital compatibility so that you can take action on the insights. GxP compliance does not only pertain to a data lake or other central repository; it is also how your data, models, and applications are governed over their life cycle – ensure that GxP is at the core of the solution in its entirety (connect and we can share more).
  • Data acquisition cannot be limited to systems data; it must be holistic to include manual operations, external environment, raw materials, utilities, and every external data that is applicable. 
  • The best and only way to understand and predict biosystems for transformative value is through the application of multivariate analyses using AI.
  • Consider the industrial scale-up of your digital twin strategy– ensure you are future-proofing for Continued Process Verification (CPV) and beyond (we can advise, design and empower your roadmap – just drop us a line).

For more information, register today for this free webinar at Xtalks now available on demand: 

Smart Manufacturing with Artificial Intelligence: A Digital Twin Strategy for Accelerated Innovation and Industrial Scale-up.

Presenting is Dr. Toni Manzano, CSO and Co-Founder, Aizon, and Luiza Mukaeda, Industry Specialist, Aizon

We hope you will join and please pass this along to your twins and/or colleagues!

By Luiza Mukaeda, Solutions Engineer & Industry Specialist

March 11, 2021

Xavier University’s AI Operations Team Re-envisions Digital Maturity

AI/ML

Many in the industry are aware of the BioPhorum Digital Plant Maturity Model first published in 2016 and updated in 2018. This model is helpful in assessing the current maturity level of a manufacturing plant and providing a common language as well as the next mileposts for IT and digital transformation teams to work towards on an organization’s journey to Pharma 4.0 and the realization of the adaptive plant. 

During the 2019 journey in the newly formed Xavier University AI in Operations Team (AIO), which I have the honor of serving on, we recognized that while the BioPhorum model is excellent at driving IT discussions, members and non-members agreed that there is more to digital maturity than digital aspects. So together, with collaboration that included the FDA, we started to envision what this model might look like specific to artificial intelligence (AI) maturity, knowing that more than the digital technology innovations would be included. We set out to release our vision at the Xavier AI Summit in August of 2020, and this vision came in the form of the AI Maturity Model Poster, which was made freely available.

The first important thing to understand when looking at the Xavier model, is that pharma companies need to already have some level of digital maturity. Consider a level 3 maturity in the BioPhorum Digital Plant Maturity Model as our starting point for our journey in the Xavier University model. The Xavier model takes into account tools and techniques like the digitization level across the company, analytics capabilities, and IT capabilities. It progresses through the maturity of data management like data quality, volume, source, structure, accuracy, and accessibility. The model continues by looking at governance and organizational factors like the ability to influence executive insights and strategic direction, Good Data Science Practices, technical capabilities and how to align to organizational goals, and how proficient teams are with AI-based tactics. Finally, the model culminates with culture and helps organizations analyze communication practices, procedures, and trust required to embrace change. The Xavier model facilitates discussions in organizations because the role of AI is a culture topic, a quality topic, a data management topic, a governance topic, and a C-suite topic.

Another crucial aspect to understand is that not all companies need to aspire to be at the top level. Organizations evaluate where they are and where they would like to be based on their goals. 

The hardest part of AI acceptance is overcoming the status quo. If a company is reluctant for innovation, they cannot adopt AI. Innovation means the desire to evolve, to leave your comfort zone. If an organization does not have this culture, then AI projects will fail. 

Need help figuring out where to start? Aizon’s AI Consulting Services can help you plan your path.

By Toni Manzano, Chief Science Officer for Aizon

Reference: AI Maturity Model Poster by Xavier University AI in Operations Team, August 2020

AI Maturity Model Poster

February 3, 2021

Smart Manufacturing and Bioprocessing – Driving More Value and Consistency from Your Bioreactors

AI/ML Continuous Manufacturing

Bioreactor processes in pharma manufacturing include some of the most evolving and complex technologies available in the production of mRNA, plasmid DNA, monoclonal antibodies, and a multitude of other vaccine types and medicinal products. The urgent demand of vaccine development and manufacturing due to Covid-19, all while maintaining controlled strategies, has placed immense pressure on bioreactor operations to increase yield with less risk and waste.

To this end, pharma manufacturing teams have been mandated with improving bioreactor process consistency, maximizing productivity, and increasing right first time (RFT) critical process parameters (CPPs). Teams also need the ability to accurately modify and control process parameters to achieve tailored products to specific requirements and the quality target product profile (QTPP).

The complexity of bioreactor data that influence these KPIs— hundreds of simultaneous parameters, poses significant analytical and data management challenges. This requires a new approach to optimize actionable steps and to alarm operators of deviations or anomalies. New Pharma 4.0 technologies are now available to gain efficiencies in bioreactors while still maintaining GxP requirements. Emerging technologies like AI, ML, cloud, and IIoT help manufacturers fully use, monitor, analyze, understand, and control the upstream manufacturing process.

Aizon is leading a webinar, “Smart Manufacturing and Bioprocessing – Driving More Value and Consistency from your Bioreactors.” This webinar will help leaders and innovators involved in achieving business value and optimizing processes based on data-driven insights. Learn how these emerging technologies can be implemented in your environments and can even integrate with classical statistics to consider all the relevant factors and variables in the system and draw insights and make predictions that would have otherwise been missed, enabling manufacturing production and quality teams to achieve their goals.

We will discuss:

  • Bioreactor yield prediction and optimization, with powerful automated AI options and phased contextualization, tuned specifically for delivering relevant bioreactor intelligence for value return
  • Deep process knowledge and batch optimization including accelerated root cause analysis (RCA) and suggested actions for batch redirection
  • Advanced anomaly detection for prediction of issues and potential failures, allowing operating engineers to react before the problems occur; scale this capability to any number of sites
  • Management of bioreactor data to preserve GxP compliance and accessibility under FAIR and ALCOA+ principles

We will also review how bioreactor advanced analytics are compatible with continuous manufacturing as a seamless path for your future Pharma 4.0 initiatives.

WATCH IT ON-DEMAND

February 1, 2021

Aizon Launches GxP AI Bioreactor Application for the Pharma Industry to Scale Manufacturing & Quality

Corporate
The application makes the power of AI accessible to pharma manufacturing while preserving GxP compliance, enabling data-driven decisions faster in the pursuit of bioreactor process optimization.

SAN FRANCISCO – (February 1, 2021) – Aizon launches its Bioreactor Application, the pharma industry’s first predictive analysis and deep knowledge management application. The Aizon Bioreactor Application adds the bioreactor process to the “Smart Manufacturing” transformation that is accelerating across the Pharma and Biotech industry. Aizon is uniquely able to achieve GxP compliance and provide an audit trail from the start of the process. The turnkey application allows pharmaceutical and biotech companies to leverage rich datasets generated during upstream manufacturing to detect and accurately predict deviations and outcomes—potentially leading to hundreds of millions of dollars in cost savings, reduced risk, and additional revenue upside.

Designed to work with both continuous and fed-batch bioreactors, the Aizon Bioreactor Application provides a deep understanding of the customer’s bioprocess manufacturing lifecycle. The application leverages specialized Edge AI and contextualization, which is a requirement to make the data actionable. Contextualized data is harnessed, analyzed, and visualized through a persona-based lens for faster root cause analysis (RCA), real-time monitoring, and predictive insight across any number of bioreactor units and manufacturing sites.

AI models for bioreactor analytics are ready for use within the application or can be configured by customers within the GxP cloud-based application. “It is clear that biotech and pharmaceutical companies need an accessible, industry-specific tool that accelerates their path to value. Customers seek a seamless way to analyze complex data sets in real-time and to predict with accuracy the yield from bioreactor units that span multiple sites and global regions,” said Pep Gubau, CTO, Aizon. “Biotech and pharmaceutical companies can now more easily leverage the power of AI/ML in a GxP environment with an end-to-end lifecycle governance of data, models, and applications in order to understand and optimize bioreactor processes in commercial manufacturing.”
Aizon’s CEO, John Vitalie adds, ”Our focus is to empower customers to innovate and rapidly achieve their targeted outcomes without the overhead burden of tracking changes, revalidation, and on-going development to maintain compliance. By building the process knowledge framework, compliance, and scalability into the application, customers benefit significantly from the streamlined industrialization of their digital solutions. We see that this is key for pharma and biotech to realize the vision of smart manufacturing and accelerate their progress toward realizing the promise of the adaptive plant.”

Pharmaceutical and biotech companies can use the Aizon Bioreactor Application to further their digital transformation journey towards Pharma 4.0 and future-proof the path to continued process verification (CPV).

The Aizon Bioreactor Application is generally available to pharmaceutical and biotechnology manufacturers today as a single application or as a seamless solution within Aizon’s Enterprise Platform. Contact sales@aizon.ai or download the data sheet for information.

About Aizon

Aizon is an enterprise software provider that transforms manufacturing operations with the use of advanced analytics, artificial intelligence, and other smart factory technologies focused on optimizing production within Pharma and other highly regulated industries. The Aizon AI platform and native GxP based applications seamlessly integrate unlimited sources of structured and unstructured data to deliver actionable, real-time insights across all manufacturing sites. Aizon brings deep domain expertise and works closely with Global System Integrators and technology partners to provide enterprise solutions to customers.

January 28, 2021

How Advanced Analytics and AI can Help Lead along the Path of Global Population Vaccination

AI/ML
Analytics + AI Path to 2 Billion Doses
The World Health Organization is leading the global charge to deliver two billion COVID-19 vaccine doses by the end of 2021. This amazing challenge includes organizing more than 150 initiatives for various vaccines from all around the world. Beyond this, there are 5 billion adults in the world and assuming most COVID vaccines will require two doses per person, that leaves us with a global vaccine dose volume approaching 10 billion.

The Challenge

There are three main challenges to overcome to achieve this scale of manufactured vaccine and safe distribution and administration. The first is managing the tremendous complexity from clinical trials to manufacturing to supply chain, all with different paths to the same goal. The second major challenge is in the scalability of manufacturing the number of doses needed to inoculate the world’s population. And third, in addition to the speed that these vaccines can be manufactured, how is prioritization decided, and at what price points. 

Global Vaccine Projections, January 2021

Table: Leading vaccines as of Jan 2021

References:

A Critical Part of the Solution

Advanced analytics and artificial intelligence (AI) are realities today. While pharma has lagged technologically for many reasons, now is the time to harness the analytical power to increase efficiency and expediency while reducing risk. The regulatory bodies have often been cited as one hindrance to progress, however, these organizations are now encouraging the use of advanced technologies to help speed manufacturing and improve the quality of drug products. This global challenge is a great opportunity to drive this change. 

We believe that access to the COVID-19 vaccines is a human right and that we can join the world in this singular global purpose, helping Vaccine Owners to cross through these challenges.

Thinking globally also means thinking about emerging countries. If we do not inoculate the entire world, the virus will continue to spread. Given the supply/cold chain issues and the volumes of doses required, it makes sense to manufacture as close to where the distribution is as possible. Here is where contract manufacturers can make the difference between global access or access only for the wealthiest countries. Technologies can be part of the solution to safeguard against counterfeit doses, ensure quality across multiple locations, and provide validation metrics allowing pharma companies to provide the best quality, safety, and effective doses that remain compliant with the regulatory requirements.

The inherent complexity of a global vaccine manufacturing and distribution requires countless repetitive actions managed by humans that could introduce errors and cause the chain to break. AI is the ideal tool to do tasks requiring human cognition without rest, with accuracy, and systematic iterations (remember – we must reach approximately 10 billion doses globally). Only AI can monitor, analyze, and predict multiple simultaneous processes with hundreds or thousands of variables. AI brings value to humans making critical decisions and operating those tasks. Industrialized AI technology to operate with GxP compliance is here, tested, and ready to help us make the world a safer place again.

Watch Aizon’s recent webinar on, “Smart Manufacturing Meets the Bioreactor: Driving More Value and Consistency from Bioprocessing.” Improving manufacturing processes will improve yield and get these vaccines and therapies into everyone’s hands faster.

January 4, 2021

How to Avoid AI Failures

AI/ML
Avoid AI Pitfalls
I was reading a Gartner research report, "How to Avoid Data Lake Failures" by Nick Heudecker and Adam Ronthal, and was struck by the similarities between the pitfalls they refer to regarding data lakes and how that translates into other technologies like artificial intelligence (AI).

Some of the highlights of their findings are, and I paraphrase:

  • Data lakes are expected to be the “silver bullet” that will address all of an organization’s data issues
  • Data lakes are implemented without specific goals in mind
  • Data lakes have technical limitations that are often overlooked, misunderstood or ignored

This all translates really well if we replace “data lake” with “artificial intelligence” or even other new technologies. Let’s start by stating clearly, artificial intelligence is not a strategy. Any time we look at new technologies, we should always start with the question, “What do we want to accomplish?” We have met a lot of people who want to put all data into a data lake to create a “single source of truth” and then want to implement AI. Then they look for a problem to solve. We recommend turning that around and starting with a clearly defined purpose.

Another problem that we see regularly is the misconception that technology is all things to all people. For example, there is often the tendency within organizations to dump all data into data lakes to keep forever and then add AI on top to solve all problems. So let’s go back to the purpose. What is the project, the context, the rules, and let’s make sure we have clear objectives defined. Without that step, the problems have just migrated to new technology.

We hear a lot of talk in the industry around the terms “digital transformation” and “Pharma 4.0” which are, at their core, all about being transformative. However, some organizations are just looking to modernize or harmonize their data environments while continuing to run existing infrastructure which represents significant technical debt. Starting with an assessment which enables the opportunity to pause and look realistically at dimensions like the amount of technical debt, and the scope and speed of the transformation they are trying to accomplish is admirable. This is where companies like Aizon, who have spearheaded the use of artificial intelligence in regulated environments, can really help by bringing an outside perspective on where a company is and what the roadmap to get to where they want to go should include.  We do this all with your need to adhere to compliance standards in mind. The benefits are compelling when we go through this kind of planning process.

If you’re interested in exploring how Aizon’s AI Consulting Services can help you to jumpstart your technology journey or help you to define that roadmap, request a consultation.

Learn more about Aizon’s AI Consulting Services by watching my video:

By Lawrence Baisch, Chief Customer Success Officer for Aizon

Reference: How to Avoid Data Lake Failures by Nick Heudecker and Adam Ronthal, Gartner, refreshed 16 December 2019

September 30, 2020

Xavier AI Summit and Workshop 2020 Reflections

AI/ML
Xavier Summit logo
I was sad to not get to see so many faces this year due to COVID, but moving to the safer, digital format didn’t mean that there weren't some great takeaways. I have my two favorites, though.

The first one was the AI Maturity Model that we created in the AI Operations Team. It was our new initiative where pharma companies can self-identify themselves to which level they are in terms of AI adoption. But not only that, they are able to describe what is the journey that they want to take in order to achieve the expected final level in terms of AI adoption. This was a big undertaking which included the FDA’s involvement as well as many pharma companies in conjunction with the AI Operations Team at Xavier University led the AI Maturity Level initiative. This model is free and available on the Xavier site. This is exciting to me because this demonstrates what happens when universities, regulatory bodies, and private enterprises come together to solve problems.

My second favorite takeaway is that AI can be demonstrated as something real in pharma now. This year, we were able to present different use cases. We saw use cases with Pfizer and with AstraZeneca. I had the pleasure to interview Matt Schmucki, the Quality Lean Coach from AstraZeneca, where he explained the process that the AI Operations Task Force did with him and his AstraZeneca team to identify critical factors and potential problems in the granulation operation at AstraZeneca. Matt said, “The results gave the team a new way to look at the data that we wouldn’t have gotten to without using AI.” He continued that, “with over a hundred different process parameters, it’s difficult to know which multivariable interaction to focus on but AI is very good at looking at these numbers and these trends.” This is incredible and wouldn’t have been possible just a few years ago.

It’s a wonderful time to be a part of AI in pharma. AI is real and we have a maturity level model that points to the journey we can follow to realize the promises this technology can bring.

By Toni Manzano, AI Operations Team Lead for Xavier University and Chief Science Officer for Aizon

Toni Manzano is a co-founder and Chief Science Officer of Aizon, a SaaS company that is transforming manufacturing operations with the use of AI and IIoT technologies for pharmaceutical and biotech companies. For over two decades, he has led software projects for international pharmaceutical companies covering the entire production process and supply chain. Today, Toni is a part of the scientific committee with PDA Europe and with the AI Xavier Manufacturing team. He worked as a researcher at the University of Barcelona as a physicist and teaches big data and artificial intelligence in postgraduate courses at the UAB. He is also a member of the Science Experts in the Spanish Parliament on big data and artificial intelligence. He has written numerous articles and holds a dozen international patents related to the encryption, transmission, storage, and processing of large volumes of data for regulated environments in the cloud.