5 principles to use when demand forecasting for pharmaceuticals

25^th January 2022

By Bevin Likuyani, Pharmacist, MBA, CSCP

In the context of pharmaceutical supply chains, forecasting can be defined as the process of predicting sales and consumption of pharmaceuticals so that they can be purchased in appropriate quantities in advance. It should be the goal of all supply chain managers to strike the right balance between demand and supply as this could have significant implications on overall success of the business. An excess supply of medicines may mean high storage costs and astronomical financial losses while demand that exceeds available supply may mean loss of current and potential clients to your direct competition. Demand forecasting is therefore a crucial aspect of pharmaceutical supply chain management and in this article, I discuss 5 principles to consider when forecasting for pharmaceuticals:

1. Focus on demand rather than customer orders

There are fine margins between what demand is and what customer orders are in the concept of forecasting. Organizations should focus on actual demand rather than customer orders. This is because customer orders are influenced by a myriad of factors that may greatly distort the forecasting process. For instance, a drug wholesaler may make an unusually huge order of Panadol tablets not necessarily because the dispenser’s demand has increased but because he anticipates a shortage of the same. In the same vein, a wholesaler may order a significantly lower amount of ibuprofen tablets just because the distributor is experiencing stock outs. In a third scenario, customer orders may be returned due quality issues. Based on these reasons it is therefore not advisable to anchor forecasting on customer orders but rather on actual demand which can significantly differ.

2. Forecasts are almost never accurate

It doesn’t matter whether you use the most sophisticated statistical techniques or the most accomplished industry experts, it is important to understand that forecasting pharmaceutical demand is almost always inaccurate. During the course of doing business, circumstances and minds change which in turn affects decisions. As someone once said, the only thing constant in life is change and therefore forecasts are just estimates. The goal is to ensure that forecast errors are kept at a minimum through constant review of forecasting techniques. Some pharmaceutical companies make the mistake of basing their forecasts on what they target to achieve rather than what is the actual demand of their products. This is a mistake. Pharmaceutical forecasts should be based on data rather than organizational goals and targets.

3. Always include an estimate of error

Building on point 2 above, it is inevitable that some form of error may result during the forecasting process. Careful statistical analysis of the variability of demand should be carried out to determine the error estimate and this should be in monetary terms. If supply chain managers notice a huge estimate of forecast error, then it is advisable to go back to the drawing board, review the forecasting steps or restructure the pharmaceutical supply chain to accommodate the uncertainty of demand your medicines are experiencing.

4. Forecast pharmaceutical classes rather than a single drug

Pharmaceutical supply chain managers should focus on forecasting pharmaceutical groups rather than a single medicine. This concept of forecasting based on groups rather than single items is referred to as risk pooling. The thinking is that, the low forecast items shall be offset by the high forecast items meaning the overall risk is lower than the average of all risks in the pool. For instance, it is better to forecast for antibiotics as a group rather than individually forecasting for amoxicillin, there is less error that way.  Alternatively, one can as well pool risk by forecasting for a group of customers ordering the same item, rather than forecasting an individual medicine for an individual customer.

5. Short-term forecasts are better than long-term forecasts

Have you ever wondered why financial institutions charge more interest in long term loans rather than short-term ones? This is because in principle, long term plans are more prone to error since they are more likely to be derailed by chance and change. For instance, someone who had a four-year demand forecast of certain classes of pharmaceuticals in 2018 may realize their numbers were mulled with error since they did not factor in the COVID pandemic which he hadn’t foreseen. It is therefore advisable to do shorter-term forecasts and do appropriate forecast accuracy reviews to reduce estimate error. Pharmaceutical supply chain managers are also encouraged to ensure short lead times as this reduces the forecasting horizon thereby increasing accuracy of the forecast.

Bevin Likuyani is an experienced pharmacist with over ten years in the industry and has an MBA from the School of Business, University of Nairobi. He is a Certified Supply Chain Professional (CSCP) from the American (Association of Supply Chain Management) the founder of Pharvers and works there as a health supply chain consultant.

bevinlikuyani@pharvers.com