January 9, 2020
Industry Calls for Improved Clinical Supply Forecasting Solution
Written by: Amy Ripston
Clinical supply and clinical operations professionals must collaborate to understand and estimate demand – especially as trials evolve. Since many facets of a clinical trial are unpredictable, clinical supply planners (CSPs) are challenged with updating forecasts and supply models on the fly. However, the tools that are meant to empower CSPs and arm their clinical counterparts with the information to make critical decisions simply do not have enough speed, flexibility, and reproducibility to keep pace.
In this interview, two senior industry clinical trial supply professionals, Thomas Møller and Charlie Yatawara, will discuss experiences with manual tools (i.e. excel) and existing commercial solutions as well as the need for an improved clinical supply forecasting solution.
Jan Pieter Kappelle (JP): What are the biggest pain points you have experienced with clinical supply forecasting within your career?
Thomas Møller (TM): Overall, decisions are made to use certain tools based on internal knowledge and resources to invest in both people, processes, and technology. Many of the commercial tools price out smaller organizations and the reliance on manual models can only take you so far. Clinical supply forecasting tools need to evolve with the complexity of new trial models, and the need for accurate, real-time information. Existing tools can’t support this, not in the way the industry needs it to be.
JP: Tell me more about your experience with manual models (Excel). There is a level of comfortability with Excel from an end-user perspective. Why did you decide this was not the best approach?
Charlie Yatawara (CY): Imagine every CSP in your organization is building their own manual model, and each in their own way. It is inevitable that there are errors in the spreadsheets, but to make matters worse, there was no standardization across the organization which makes it very hard to have conversations with internal stakeholders. Additionally, it was scary to have the knowledge of a specific individual. What happens when they go on vacation? Or left the company? Not the best solution for business continuity.
JP: Understandably so, you ended up moving to an existing commercial solution. Can you walk me through the experience of going from manual spreadsheets to a tool?
TM: While at a previous organization, we implemented a commercial tool. We were lucky. We had both the internal resources and SMEs in-house to build a tool that would sit on top of the commercial tool giving us the ability to pull in real-time actuals and make decisions off that data. We had the ability to make decisions on a high-level view of the data.
JP: Building a homegrown tool must have been quite an endeavor. Can you elaborate?
CY: The entire process took the better part of a year, if not longer. As Thomas mentioned, we had the resources and the SMEs in the building to drive the process, but it took a lot of man-hours and the collaboration of KOLs and developers to get the inputs needed. In the end, it had everything we needed from an operational perspective upstream.
JP: Can you explain the frustrations of using the commercial tool?
CY: Primarily it came down to speed and complexity. If the study was too big, it took too long to run simulations or in some cases wouldn’t run at all. To provide some context, it could take 20-30 minutes to run a model for a small trial. The decision then focused on how many runs you could execute. While you may have wanted to run 100 to have better data, we may have settled on 10 because it was taking too long. We would avoid anything over one hour, but our true pain point was 20-30 minutes. We would lower the number of runs, and therefore the accuracy, test to see if the model works, then take it from there. It was a lengthy, cumbersome process.
JP: So, do you agree there is a need for an improved forecasting solution? What does that look like to you?
TM: Coming from our experience, utilizing both Excel and commercial tools, we are brutally aware of the limitations of both approaches. We need a tool that will do the job, that lets our whole team work in a unified (faster and more user-friendly) way. Commercial tools are too much of a heavy lift on internal knowledge. We also are looking for a tool that is fully cloud-based, that we can use no matter where we are and not be tied to our physical computers.
CY: We have been very fortunate to collaborate with 4G Clinical to inform the next generation of supply chain optimization. It has been an investment in our future, and there is good energy and innovative spirit to the entire company. We look forward to tackling these challenges together.
Tag(s): Supply Optimization , Innovation , Supply Forecasting