Bluefin

Let’s talk site resupply strategies. They’re important, they’re complex, and they’re often something of a dark art. Bluefin makes it automatic and scientific: we analyze your study and recommend optimal RTSM predictive resupply strategies and site buffers. As a clinical supply manager, you know how complex and dynamic RTSM site settings can be. They need to account for higher- or lower- enrolling sites or countries, lead times, and even local sourcing. And they’re dynamic: as the study enrolls, the optimal buffers and predictive windows change. This can happen as frequently as every week. Distribution planning may not get as much attention from outside stakeholders as demand forecasting, supply plans, and expiry extensions, but clinical supply managers know that doesn’t make it any less important. Interested in seeing RTSM supply recommendations live? Get in touch. #rtsm #clinicalsupply #supplychain

Exciting news: tomorrow our CTO, Chris van der Loo, will be speaking at Clinical Trial Supply New England at 4:30 pm ET. Join us as he talks about how digital twins improve clinical supply forecasting, and shares three real-world case studies. CTS New England is free to attend for sponsors, and takes place at the Westin Waltham. https://lnkd.in/e74Qu-UC

Whenever I come to the Bay Area in California it feels like I’m living in the future: the driverless cars, the tech company billboards advertising things I’ll hear about in a few years. The future really is HERE, it’s not evenly distributed yet. And we’re discussing the future of clinical supply chain here at #BSMA as well. But that’s not to say that’s all we’re talking about. We just heard from a patient advocate who is fighting cancer. And then there was this…

Want to reduce shipments and end-of-study waste? Watch this short clip and see how Bluefin enables you to adjust depot buffers and instantly see the impact.

In Bluefin v4.3, we've added visuals indicate where lots (both existing and recommended) are being allocated. We won't lie: we've wanted to use a Sankey diagram for this for quite a while...

Bluefin has kicked off 2026 with version 4.0 that sets the tone for where we’re headed this year. More optimization-driven recommendations. A powerful forecasting engine. And visualizations that help you make real operational decisions, quickly and easily. A few highlights in 4.0: • We've added distribution recommendations such as depot buffers and site resupply settings to our existing mathematically optimal production suggestions • An even more powerful forecasting engine with details such as DNX dates - things often approximated in spreadsheet models and other planning software, but that make a real difference in preventing stockouts We’ve never been more excited about our roadmap and the benefits Bluefin brings to clinical supply teams.

We hope all our friends in clinical supply are taking some well-deserved time off over the holidays. But clinical trials and patient visits often don't work around holiday schedules. Clinical supply teams: how do you prepare for the break? What tips, tricks, and best practices can you share that may help help others?

One of the features we’re asked for the most is the ability to run alternate scenarios. A lot of people call them “what-ifs.” Creating and organizing many scenarios can be one of the most challenging feats to achieve in spreadsheet models. And we have listened. In Bluefin, you can model an unlimited number of alternate forecasts. Bluefin lets you create alternate scenarios instantly by cloning 👬 (trying to use emojis here, ok?) your existing scenarios. From there, you can tweak as needed. And keeping them updated, in sync, and organized is effortless. Some of the most common use cases that we hear from customers: 🏎️ 🐌 Faster or Slower Enrollment As a supply manager, you’ll start off with a baseline forecast from your CRO or Clin Ops team. Often, these forecasts prove to be optimistic. So, you want to model a… shall we say… alternate reality in which enrollment is a bit slower than anticipated. You need to know how much product will expire sitting on shelves. But the opposite is also possible! You may also wish to model a scenario in which certain, or all, regions enroll faster than anticipated. Of course, in this case, you need to make sure you have enough at each depot to fulfill demand. The ability to model multiple scenarios helps meet the demands of the study while reducing waste through optimized lot sizes and timing. 💭 Conceptual Study Design During study design, you may be asked to generate top-down forecasts for wildly divergent study designs. “What if we add an arm?” “Change dosing schedules?” “Delay enrollment by a year?” “What if we’re in XYZ regions?” Bluefin helps you take these questions in stride and model different designs with speed and agility. 💊 And So Many More There are so many situations where being able to quickly assess alternate scenarios can save the day. Here are just a few more of the challenges our users are often faced with. * Mitigating short shelf life and expiry extension * Optimizing waste reduction by testing various buffer strategies * Changes in regions, countries, or sites * Visit schedule or protocol amendments * Formulation changes Let’s set up some time to talk about your studies see how Bluefin can help you with your forecasting. Message us on LinkedIn, or set up a demo via our website at www.bluefin.one.

While at Clinical Trial Supplies East Coast last week, I joined the roundtable discussion led by Alina Borovika on using AI in clinical supplies. A couple of really sensible questions came up, generally along the lines of: "can't I just throw my protocol and all my enrollment and inventory spreadsheets into my enterprise version of Copilot and have it forecast and plan supplies for me?" My reaction: give it a shot; it might surprise you! It turns out one of the people had already tried this, and Copilot struggled to even parse the protocol correctly -- let alone the data on things like inventory. And the challenges go beyond what could be remedied with prompt "engineering." Why might this be? Are LLMs simply not quite "there" yet? Or are LLMs not the right tool for the job? Perhaps some of both. Generally, I feel that LLMs excel where there is a lot of open source text they were trained on; there are a modest number of clinical trial protocols available, but it's not huge. Another area where forecasting likely plays to an LLM's weakness is the requirement to understand and then devise calculations to perform on the relatively difficult-to-make-sense-of data from various spreadsheets. I think it's totally reasonable to ask why we can't just drop a bunch of files into an LLM, tell it what we need, and get a result. It looks like right now, dedicated tools (like Bluefin) which are engineered with domain-specific knowledge are still the winner, and by a wide margin.

We're excited to return to #CTSEastCoast on Tuesday and Wednesday. Come by our booth and see the latest in clinical supply planning technology, including our new production recommendations that take into account things like expiry extensions, country labels, and other constraints. These opportunities to spend time with the broader clinical supply community discussing today's challenges and the state of trials and technology always motivate us and drive us forward. See you there!