Amaury Jeandrain
Which AI Method for Clinical Supply?
[EPISODE 2] - Simulations
Can you imagine being able to look into the future and foresee what might happen? And what would happen should you make different decisions?
This is what simulations offer. Simulations are essentially an imitation of the operation of a real-world process. It replicates what could happen in the future, making a process that would take years in real life, happen virtually in a matter of minutes.
Simulations are therefore often used for predicting and analyzing complex events, visualizing the long-term impact of decisions, testing scenarios, as well as performing risk analysis. They are crucial for modeling complex systems and environments (which defines the clinical supply chain well enough), and are a significant part of AI research and application. One of the main strengths of simulations is their ability to consider a wide range of variables as well as their interdependence.
Simulations can be confused with digital twins, due to their capacity to generate a digital version of the future of a clinical trial. But while simulations replicate what could happen in the future, digital twins replicate what is actually happening in the real world.
What are the applications of simulations in clinical supply chain management?
The applications of simulations in clinical supply chain management can take place in four areas:
Let's see how exactly.
1. Forecasting and prediction
Simulations can provide highly detailed demand forecasts. These forecasts exceed the accuracy of deterministic models due to the comprehensive nature of the simulations, which can account for a wider range of variables and interactions. As an example, the number of patients recruited per site tends to be very low in clinical trials, sometimes less than one, and that one patient might change doses over time. In this case, decisions made based on averages would be catastrophically risky or offset by overly conservative safety buffers.
In the clinical supply chain, using simulations can be very powerful in accurately understanding the patient’s treatment possibilities (aka, each patient would virtually be simulated: his recruitment, screening, treatment, titration, drop-out, drug dispensing), while considering the inherent uncertainty of clinical trials.
2. Planning
The supply chain can also be simulated accordingly, with each kit being packed, labeled, distributed and dispensed to respond to the simulated demand. A simulated supply chain is an excellent basis to optimize a supply chain strategy (we will talk about optimization in our next episode). Simulations also offer additional benefits in understanding the relationship between multiple variables (e.g expiry events and waste, protocol design and costs).
Another advantage is that simulations can also provide very detailed supply chain forecasts (such as inventories, shipments, costs, emissions, etc.) that can support the planning of the supply chain.
3. Decision intelligence & recommendation systems
While forecasting & planning can not be generated by Generative AI, decision intelligence and recommendations is an area where interactions between GenAI and simulations can be very powerful.
As simulations would make a trial happen from beginning to end, they offer the opportunity to test different versions of the trial. What if we implemented lot pooling? What if the protocol design was different? What if we sourced at a different frequency? For any question, any decision, any event, simulations give highly accurate impact assessments.
Simulations can therefore be used by both leaderships teams and clinical trial teams as they allow seeing the outcomes of any option without actually having to run the clinical trial. On the one hand, leadership teams will use them to make strategic decisions (e.g. should we close a depot, should we invest into digital labeling or lot pooling). On the other hand, clinical trial teams will use them for trial and supply chain design purposes (kit design, network design, protocol design, country selection, etc.).
Simulations are thus often used to train AI models (e.g. to generate synthetic data for deep learning). But a well built GenAI model could also interact wonderfully with a simulation engine, as a Copilot, and recommend scenarios to simulate. But GenAI will never be able to predict outcomes accurately, like a simulation model can.
4. Risk analysis
Risks are inherent to clinical supply chain management, just as much as clinical trials are unpredictable. Through scenario testing, supply chain managers can perform sensitivity analysis on clinical assumptions (recruitment, titrations, drop-out) to plan for a risk-free supply chain. They can also assess the impact of constraints (delays, shortages, temperature excursions) to assess future risks and stress test their decisions (closing sites, shipping under quarantine, bringing back drug from a depot, etc.) to identify coherent mitigation strategies.
Here too, a GenAI model trained by simulations could easily counsel supply chain managers on which scenarios to test, their likelihood, as well as mitigation ideas. It would save a lot of time and effort for users and make the simulation process lighter, but they would always need to simulate those ideas to have an impact assessment.
Other AI methods for clinical supply chain management
In the next episodes of this series, we will analyze how other AI methods can support the clinical trial supply chain. Look forward to our analysis on the following topics in the coming weeks:
AI in Clinical Supply Chain Episode 3 : Optimization
AI in Clinical Supply Chain Episode 4 : Machine learning
AI in Clinical Supply Chain Episode 5 : Other methods
Want to receive the next article directly in your mailbox?
If you missed it, you can find the first episode of this series here:
AI in Clinical Supply Chain Episode 1: Generative AI
AI at N-SIDE
As an optimization company, N-SIDE has been successfully using AI for more than 20 years in over 12.000 clinical trials. Optimization, simulations and machine learning algorithms are native in our solutions.
Wish to learn more about how N-SIDE uses AI to reduce drug waste and risks in the clinical supply chain, while saving time and money? Book a meeting with one of our experts.
