Forecasting has always relied on large volumes of data. With the complexity of data generated daily across different data sources, leveraging this effectively to improve the accuracy of forecasts is even more crucial in supply chain planning. Understanding how to improve your forecasting accuracy with the wealth of information available is key to successfully managing any supply chain, and the pharmaceutical industry is no different.
There are several reasons why your forecasting may fail you. A lack of relevant training and inaccurate risk assessments are among the main causes. And while there is a constant battle for an accurate forecasting system, companies are faced with the economic need to decrease drug waste while saving money. While there are many forecasting tools and software out there yet to be explored, there is no counterbalance for a lack of understanding of supply chain risks.
Decreasing drug waste and causing actual decrease in manufacturing creates a reactive rather than active approach to your supply chain control, leaving you to additional ad-hoc manufacturing of drugs as per need. Certainly, there are cases when the drug is delayed already due to several issues related to shipping for instance. International shipping can be a nightmare. It is very complicated to coordinate all the parties taking part in the shipping process. First of all, there is the customs broker and the sponsor company, then the import companies. For example, there is no universal agreement even on who should do the importer record. Furthermore, the FDA regulations on shipping can at times be quite vague and there are times when supplies are often held at borders. This is especially the case in countries like Brazil, Argentina or Russia, which can be challenging. And to make things even worse you must deal with additional players in the shipping processes like the distribution vendor. In addition, all of the above mentioned parties may have separate procedures and documentation needs.
Typically, then the sponsor companies have different affiliates in the destination country. If so, this would make things easier, as once you have an affiliate after the shipment arrives to the country they can take over the shipment and ensure its delivery to the sites. If there is no affiliate you need to use another company that provides you these services, but of course many times you may have no idea about their credibility and quality. With shipments it is always complicated. In countries where political instability is inherent, you may be forced to reconsider the logistics of your shipment to avoid hold-ups and substantial delays. Additionally, high taxes in foreign countries can sometimes leave you way over budget, so be mindful.
Furthermore, sometimes the patient enrolment slows down where you can't get enough patients enrolled for the trial on time, meaning you face time constraints, which ends in significant drug waste. Clinical trials are very dynamic and often you can't forecast them properly, but having 5 percent overage is not enough. 30 percent may still be a silly number to go with and would need to increase it dozen more times.
Obviously, all the issues mentioned above cause your company extra costs and hence forecast accuracy is critical to the bottom line. However, many companies still rely on subjective forecasting methods with very little data to back them up. In fact many professionals agree that demand planning KPIs, such as forecast accuracy and forecast bias should be are the forefront of an SOP process. As companies become more responsive to patients, longer supply chains that extend overseas to suppliers and contract manufacturers mean that they have less direct control over operational processes. This makes them more vulnerable to delivery problems, quality issues, missed deadlines, inventory liability, and forecast accuracy. In most companies there exists a specific type of variability, and that is either variability in demand or, more often than not, variability between forecast and actual demand. Therein lies the importance of understanding forecast accuracy when working towards a best-in class S&OP process.