An international healthcare company ranked in the world’s top 20
Design and implementation of a simulation/optimization software for the Supply Chain Management of Clinical Trials.
Many challenges await those brave enough to take charge of the supply chain for clinical trials. Demand is highly unpredictable and associated costs and risks are high. Failure to supply a drug at the right time, in the right amount and the right conditions is unacceptable for the patients who put their health at the mercy of science, and for the healthcare company that may lose millions of dollars due to time-to-market delay. Furthermore, each clinical trial displays a unique set of characteristics, implying that it is impossible to develop stable, steady-state, optimal process such as other goods production industries strive for.
One particular healthcare company quickly understood that in order to provide a high service level while keeping their costs down, commonly used software technologies were no match for the complexity at stake.
Business experts teamed up with statisticians and called N-SIDE in order to design and implement specialized software that was visionary at the time and would enable them to:
estimate the rate of patient enrollment in the trial
estimate demand generated by patients through the multiple treatment visits
measure the uncertainty and implied risks related to this demand
infer smart warehouse management strategies to minimize costs
measure the performance of a given strategy in terms of costs and service level
N-SIDE took up the challenge and implemented a prototype simulator/optimizer. Through the years of continued close collaboration between N-SIDE and the healthcare company, this embryo progressively evolved into the most effective dedicated software available on the market: CT-FAST or the Clinical Trial Forecasting and Simulation Tool.
The number of uncertainty sources and their trial-dependent characteristics make it very difficult, if not impossible, to provide a robust, accurate, analytical supply chain model in the context of clinical trials. Yet gauging the risk is vital to ensure high service levels. The solution lies in applying Monte Carlo simulations, which enable the evaluation of complex/unknown distributions and indicators, based on a high number of random events.
We put this into practice as a prototype in 2003, to simulate the overall supply chain, from the enrollment and treatment of patients to the production, shipping and administration of the drug on trial. Analysis permitted the identification of cost-optimal strategies by balancing the shipping costs with the manufacturing costs for over-age materials that cover demand uncertainty.
Through a sequence of projects with our sponsor pharmaceutical company, the prototype became an actual product, and important functions were added: a user-friendly, intuitive graphic interface, the ability to leverage real-life data to initialize simulation and identify statistical distributions, and the ability to leverage parallel computing on a large grid of machines, in order to speed up computation, opening the door to larger, more accurate simulations.
In 2008, our client transferred the intellectual property back to N-SIDE, and CT-FAST became a commercially available software.
The key benefit for pharmaceutical companies is probably the standardization of the clinical trial supply management methodology through the use of advanced, dedicated software and standard reports. This standardization enables enhanced communication and improved process control, where individually-designed spreadsheets or back-of-the-envelope calculations were once in common use.
Another major benefit is the increased efficiency in supply chain management, through accurate simulations and fast comparison of multiple strategies. Compared with initial “guesstimates” from study managers, this new approach usually delivered savings up to millions of dollars solely on manufacturing and shipping costs!
Also, leveraging real-life data within simulations enabled a better anticipation of the kind of issues that often arise when assumptions change (patients come in faster than planned, production batch cannot be delivered on time…), which promotes strategy and foresight rather than firefighting. In particular, successful anticipation of drug shortages can yield savings of millions of dollars by avoiding any impact on the time-to-market.