Projection factors for forecasting product demand
View Patent ↗Projection factors are used to project product sales data from sample retail outlets to the universe of outlets. Weekly forecasts of market conditions and product demand are generated based on the projected product sales data. The projection factors and the weekly forecasts are updated during the course of the forecasted week, for example, daily, as data on actual product sales during the forecasted week is received.
1. A method implemented on a computer arrangement including a processing section for forecasting market demand for a product in a universe of stores over a subject time interval, the universe of stores comprising sample stores and non-sample stores, the method comprising:
generating a prediction of market demand for the product in the universe of stores for the subject time interval based on historical data on purchases and sales of the products in the universe of stores,
wherein generating the prediction comprises applying projection factors to data on purchases and sales of the products at sample stores to estimate data on purchases and sales of the products at non sample stores;
receiving data from reporting stores on actual purchases and sales of the products during the subject time interval;
calculating, using the processing section, an adjustment factor by comparing actual and predicted purchases and sales of the products at reporting stores during the subject time interval;
applying the adjustment factor to estimate actual purchases and sales of the products at non-reporting stores during the subject time interval; and
updating the prediction and projection factors based on the actual and the estimated actual purchases and sales of the products during the subject time interval,
wherein the sample stores include the reporting and non-reporting stores.
2. The method of claim 1 further comprising subjecting data received from reporting stores on actual purchases and sales of the products during the subject time interval to quality control routines designed to screen for unusual events.
3. The method of claim 1 wherein calculating an adjustment factor comprises calculating an adjustment factor based on the difference between actual and predicted purchases and sales of the products.
4. The method of claim 1 wherein calculating an adjustment factor comprises calculating an adjustment factor at one of a product level, a local geography level, a product dosage level, and a prescriber level.
5. A system for forecasting market demand for a product in a universe of stores over a subject time interval, the universe of stores comprising sample stores and non-sample stores, the system comprising a processing arrangement configured to:
generate a prediction of market demand for the product in the universe of stores for the subject time interval based on historical data on purchases and sales of the products in the universe of stores, by applying projection factors to data on purchases and sales of the products at sample stores to estimate data on purchases and sales of the products at non sample stores;
receive data from reporting stores on actual purchases and sales of the products during the subject time interval;
calculate an adjustment factor by comparing actual and predicted purchases and sales of the products at reporting stores during the subject time interval;
apply the adjustment factor to estimate actual purchases and sales of the products at non-reporting stores during the subject time interval; and
update the prediction and projection factors based on the actual and the estimated actual purchases and sales of the products during the subject time interval,
wherein the sample stores include the reporting and non-reporting stores.
6. The system of claim 5 wherein the processing arrangement is configured to subject data received from reporting stores on actual purchases and sales of the products during the subject time interval to quality control routines designed to screen for unusual events.
7. The system of claim 5 wherein the processing arrangement is configured to calculate an adjustment factor based on the difference between actual and predicted purchases and sales of the products.
8. The system of claim 5 wherein the processing arrangement is configured to calculate an adjustment factor at one of a product level, a local geography level, a product dosage level, and a prescriber level.
9. A non-transitory computer-readable medium for forecasting market demand for a product in a universe of stores over a subject time interval, the universe of stores comprising sample stores and non-sample stores, the computer-readable medium having a set of instructions operable to direct a processing system to perform the steps of:
generating a prediction of market demand for the product in the universe of stores for the subject time interval based on historical data on purchases and sales of the products in the universe of stores,
wherein generating the prediction comprises applying projection factors to data on purchases and sales of the products at sample stores to estimate data on purchases and sales of the products at non sample stores;
receiving data from reporting stores on actual purchases and sales of the products during the subject time interval;
calculating an adjustment factor by comparing actual and predicted purchases and sales of the products at reporting stores during the subject time interval;
applying the adjustment factor to estimate actual purchases and sales of the products at non-reporting stores during the subject time interval; and
updating the prediction and projection factors based on the actual and the estimated actual purchases and sales of the products during the subject time interval,
wherein the sample stores include the reporting and non-reporting stores.
10. The computer-readable medium of claim 9 comprising instructions operable to direct the processing system to subject data received from reporting stores on actual purchases and sales of the products during the subject time interval to quality control routines designed to screen for unusual events.
11. The computer-readable medium of claim 9 comprising instructions operable to direct the processing system to calculate an adjustment factor based on the difference between actual and predicted purchases and sales of the products.
12. The computer-readable medium of claim 9 comprising instructions operable to direct the processing system to calculate an adjustment factor at one of a product level, a local geography level, a product dosage level, and a prescriber level.