Supply Chain Planning for (Bio)-pharmaceuticals: A novel prescriptive analytics approach

Aktualisiert: 11. Aug. 2019

Shelf life constraints constitute a major challenge for successful #supplychain management in #pharmaceutical industries. Expired drugs result in significant costs for drug returns, destruction, and write-offs and might cause supply shortages. To reduce the risk of #obsolescence, we developed a novel data-driven approach for optimizing batch sizes and inventory deployment.

Demand variability causes drug obsolescence

Every supply chain professional would agree that forecasts are always wrong. For example, AT Kearney reports an average accuracy of 61 % to 74 % for 3-month-ahead forecasts at the SKU level in the pharmaceutical industry. Forecast errors are much more costly in supply chains with significant shelf life constraints. Overestimating demand is not simply a matter of building up inventory, but can cause obsolescence if products remain in stock too long.


A new approach for batch size optimization

To overcome issues affecting standard lot size models, we developed an extension of the well-known EOQ model. The extension takes forecast errors, shelf life, and the cost of obsolescence in addition to inventory costs and production or ordering costs into account when determining lot sizes. In addition, it allows considering the impact of artwork changes on replenishment decisions. Technically, a so-called stochastic optimization problem, which explicitly considers demand uncertainty when optimizing lot sizes, is solved.


Batch Size and Inventory Optimization in Pharma
A data-driven analytics approach to optimize batch sizes and replenishment decisions.

Cost-risk curves and end-to-end supply chain cost

Besides calculating the optimal batch size, the tool allows organizations to review the positioning of their current and optimal supply chain strategy. For each SKU, a Cost-Risk Tradeoff Curve can be calculated to visualize how batch sizes affect cost and risk exposure. In the figure below, each point on the curve corresponds to a specific batch size. The figure shows cost and risk of the current batch size (or any other benchmark value) and the optimized batch size on that curve.


Cost Optimization and Risk Management for Inventory in Pharma
A Cost-Risk Tradeoff Curve visualizes how batch sizes affect cost and risk exposure

Our discussions with supply chain leaders confirmed that these tradeoff curves help to better align supply chain strategies and objectives across different functions such as supply chain, logistics, manufacturing, and local affiliates. In addition, such alignment is further facilitated by a breakdown of total cost into different cost categories.


Industry cases show substantial savings

The method was applied to drug products of two top-20 pharmaceutical companies to study the benefit of including shelf life constraints more appropriately into supply chain policies. The figure below shows the results of optimizing two major brands (biosimilars) of one of the pharmaceutical companies. Optimization of batch sizes and replenishment decisions with our tool identified savings up to 3 million euros per year. Furthermore, by optimizing replenishment at SKU-level, our tool helped to reduce supply risks.


Inventory and supply chain optimization for pharma and biotech
Our approach identified >3 million euro savings opportunity for two biotech brands

Summary

We developed a data-driven approach for integrated lot size and safety stock optimization. It can be used to determine batch sizes and order cycles for warehouses and packaging plants. Our tool is available as a cloud-based supply chain application or as a service. It relies primarily on data commonly available in ERP and APS systems. Industry cases show substantial cost savings compared to the supply chain-planning approaches commonly used by pharmaceutical and biotech companies.


Please contact us if you are interested in a demo or pilot study. A scientific article about our work is published in Pharmind and is available here.