Demand Forecasting for the Indian Pharmaceutical Retail: A Case Study

Abstract

A retail organizations ability to forecast demand accurately is requisite to identify, streamline and optimize business decisions. The optimization of order quantity, stock level, or delivery schedule depends on the aptitude of a retail operator to forecast accurately, demand at the store level. Forecasting is important in the perspective of the pharmaceutical industry, which commonly employs Price War tactics and requires efficient Supply Chain Management (SCM). Apollo Pharmacy is one of the largest retail chains in India with over 70 round-the-clock outlets. It is moving towards enabling e-prescription services to the consumers. This study was conducted to determine the accuracy of statistical forecasting techniques used in inventory planning of Apollo Pharmacy of the Apollo group in an Indian community at brick level. The retail outlet reported out-of/excess stock on certain business days. Therefore, a gap analysis was carried out and it was understood that information from customers to supplier is not captured. According to preliminary surveys, forecasting does not factor in their SCM process and there is underestimation of the importance of forecasting as a part of SCM. Pharmaceuticals are subject to a high degree of demand fluctuation (based on seasonality and duration of ailment, pricing etc). This article demonstrates the application and screening of simple forecast models (Moving Average, Exponential smoothing, Winters Exponential) to forecast demand for pharmaceuticals. Two products were empirically chosen for the study – Okacet 10mg tablet (seasonal demand, contains Citrizine) and Stamlo Beta tablet (non-seasonal, contains Amlodipine 5 mg & Atenolol 50 mg). Accuracy assessment parameters indicate that the least sales forecast error for Okacet (seasonal) was obtained using WES (a=0.2, ß = 0.1, ? = 0.01, ET = 92.28) and (MAPE = 27.50) indicated highest forecast accuracy, whereas for Stamlo Beta (non-seasonal) WES provided no additionally superior forecast as other models.

Keywords: Time Series, Winters Exponential Smoothing, Demand Forecasting, Pharmaceuticals, MAPE, Retail Pharmacy, SCM

View PDF