WORKLOAD FORECAST ALGORITHM OPTIMIZATION FOR RE-ORGANIZING RETAIL NETWORK

This research is focused on developing an innovative approach for optimizing workload forecast algorithms in point of sale for retailers; this paper proposes a real case as validation framework and the procedures for optimizing and fine tuning the predictive algorithms for improving their performances. The analysis is based on different time series (i.e. sales, customers, working hours, etc.) correlated by the predictive algorithms. The paper proposes a metrics devoted to measure the performances considering the multivariable framework and the different target functions.