The Background
Organizations that import and distribute prescription drugs operate under volatile environmental and epidemiological conditions that can impact consumer demand. They are also exposed to suppliers’ lead time uncertainties as well as high minimum order requirements.
Furthermore, inefficiencies in regulatory frameworks such as registration of prescription drugs and import licensing can cause disruptions in the supply chain.
To mitigate all this, such organizations typically maintain a very large inventory, sometimes as much as tens of millions of dollars. Apart from the significant working capital requirement, this model leads to large scale expirations.
Our Customer
Our customer is a very large pharmaceutical importer in South Asia responsible for sales, marketing, and distribution of over 1000 prescription medicines. Their manual inventory planning was overburdened due to volatilities in demand, supply, regulatory requirements, and the size of the product portfolio.
As a result, the customer was facing both excess inventory concerns & out of stock issues
The Approach
Leveraging accurate forecasting at scale: Due to FORECAST²’s ability to generate customized forecasts, the customer started generating accurate demand forecasts for the entire product portfolio.
Agility in forecasting: The customer started updating the existing demand forecasts whenever changes in pricing and bonus quantities occurred. Hence the forecasts always remained current and reliable.
The input of supply parameters: Product parameters such as MOQ, Lead time, Shelf life, registration expiration dates were injected into the inventory planning module for the entire product portfolio.
Continuous streaming of inventory data: Information on current inventory and orders in transit was continuously fed in to the inventory planning module.
Use of auto-order generation: The inventory planning team started using the auto-order generation function which generated the optimal order plans using Machine Learning algorithms.
The result
82+%
average monthly demand forecasting accuracy
102 days
average inventory days for the portfolio (9-day safety stock level) Zero out of stock situation
5%
reduction in inventory value
Conclusion
The customer was able to start forecasting the demand in an accurate and reliable manner for the entire product portfolio There was a better end to end visibility of future demand, supply parameters, stock levels, pending orders in one system Over time, they migrated from manual order planning to machine learning-driven order planning recommended by the system This transition reduced the planning time from weeks to a couple of days. Hence inventory planners were able to focus more on delayed orders and supplier negotiations It also eliminated the human error that can occur due to complex and extensive manual work.
Dhrmapala Mawatha,
Colombo-07,
Sri Lanka.
#27-00,
Republic Plaza,
Singapore, 048619.
Jl. Raden Saleh Raya,
No. 46A, Jakarta,
Indonesia.