||A nationwide chain of 800 service centers
||Retail store managers
||At this chain of service centers, store managers were responsible for making daily staffing decisions to ensure that each store had enough staff to meet customer demand and ensure short waiting times. However, having too many extra staff led to excess labor costs. Thus, optimal staffing required accurate, daily predictions of customer demand for each store.
||Stottler Henke developed a case-based reasoning system that predicts the daily sales volume for each of the 800 stores in the chain. This prediction tool outperforms daily predictions previously made by store managers, thereby reducing labor costs while maintaining high levels of customer service. This system also outperforms forecasting systems based upon statistical methods.
||Based on the system’s predictions, each retail outlet makes daily staffing decisions to minimize both customer waiting and staff idle time. The tool relies upon electronic point-of-sales (POS) data and does not require store staff to carry out extra onerous data entry.
||This prediction technique can be used for predicting daily, or even hourly, sales revenues for the near future for chain service businesses such as restaurants or automobile service centers.