IT/OT Convergence - Part 4

Part 4 – What Needs to be Done

The first step in determining what strategy to take in the digitization and connection of your systems is evaluating the current state of IT and OT systems and determining what changes are needed to facilitate integration. This is sometimes best undertaken by an external set of eyes that can bring in the best practices and trends from the industry as well as remove any bias from years of “being in love with the system” or a sense of being resigned to “this is how it has always been” and just getting by and keeping the lights on. In building this strategy, the executive sponsors must see the value that will be provided by this significant investment. They must be engaged as early as possible to ensure efforts do not fizzle out after the first milestone or delay. And storytelling on the “why we are doing this” and demystification of jargon is key here – blinding stakeholders with a lot of with IT/OT jargon will likely result in confusion and unmet expectations.??If “IT/OT geeks” are speaking about IOT buzzwords than their stakeholders, the MS&T geeks, may glaze over and start thinking about glycosylation of the monoclonal antibody they are working on, and no-one is any wiser after the meeting!

The next step is identifying the systems and data that need to be integrated and prioritizing the highest return investments. This needs to be done at the micro level, identifying obsolete systems and equipment, but also at the macro level. You must ask which manufacturing facilities will benefit most from this modernization in the mid to long term and not just apply the same template to all manufacturing modalities and facilities. It must also be said that the “low hanging fruit” is rarely the one that provides the most benefit. A Chemical API facility with excess capacity to execute these projects but has limited patent protection and a smaller pipeline will not provide the same game changing improvements that a slight increase in yield/OEE or reduction in cycle time in a Biologics or Vaccine bulk can potentially provide. ?Additionally, if an IT/OT transformation program can take the “high ground” and prove that they can add value and reduce risk in critical and complex nodes in the network (such as launch site for new modalities) it will be easier to bring diverse stakeholders from other functions (quality, reg affairs, finance) on board regarding the risk/benefits of the investment.?This will make it easier to “roll out” the program in the wider network.??But a transformation executed at a relatively simple site (such as a regional packaging site) is likely to be met with some scepticism from stakeholders elsewhere in the company (“yes, but our site is far more complex…”

The core of the execution is in the development and implementation of new IT and OT systems or modification existing systems to facilitate integration. The cost and effort here is obvious, obsolete systems must be replaced and upgraded in tight shut down windows and equipment and experienced resources become ever more expensive. The key here is that the cost to replace or upgrade will never get cheaper as systems and regulations become more complex, and wages continue to increase for a shrinking engineering resource pool. Even though the technology might become cheaper, and offshoring becomes more common, there are limits to what can be done remotely and systems will need to be refreshed more frequently to address security concerns. It is critical to initiate this process as early as possible in the lifecycle of the facility to reduce future upgrade and maintenance costs.

While testing and validating the integrated systems has always been a burden that the Life Science industry has had to contend with, the FDA and regulators are beginning to recognize the value of risk-based testing and validation. This is made clear in the recent guidance from the FDA (Computer Software Assurance for Manufacturing, Operations, and Quality System Software) that they do not want a mountain of validation documents slammed in front of them during an audit. It is important to conduct a risk assessment and formally document those processes and systems that are critical or elevated risk, while informally testing those which are lower risk. This is not an excuse not to test your systems but applying critical thinking to the documentation process and focusing on those aspects and systems that are most risky to the patient.

Training existing and new employees on the use of the new systems cannot be underestimated. If we put in all these complex systems, but do not effectively manage the change within the organization and provide sufficient training, it is all for naught. You will continue to have huge volumes of data, but no one that can utilize it for the benefits which it was intended. A comprehensive super user and end user training strategy is essential to maximize the return on these modernization investments.

Finally, a long-term strategy for maintaining and updating the integrated systems over time must be well thought out and properly resourced to support business as usual operations. While outsourcing can be a legitimate cost reduction initiative, OT systems have much different Service Level Agreements (“SLAs”) than traditional IT systems and internal, on-site support may always be required at some level. It is also important to keep the upgrade and patching schedules aligned with the business as 30 minutes of downtime for systems that operate in microseconds, 24 hours a day, 7 days a week, is completely different than in a website or email server.

Those that already have these foundational systems and IoT ready devices have a leg up but cannot rest on their laurels quite yet. All the data that is being generated by these systems, now needs to be collected, contextualized, and analysed. This contextualization is particularly important as many of the IoT devices and simple HMIs and PLCs have limited batch context available and may require significant effort and resources to ensure accurate and meaningful data analysis to segment time-based data into batch-based contextualized data. Further complexity comes in when one considers the batches of drug substance that might contribute to different batches of drug product.

In the final article we will look to the future and what it holds for IT/OT Convergence in the Life Science Industry.