Reducing uncertainty in demand for blood

This study addresses some of the above challenges faced by blood centers by introducing practical methods for accurate blood demand forecasting, which will allow for lowering of costs, reduction of blood wastage, and conservation of limited resources. Six blood centers were visited in New York State during the years of 2010 and 2011. In all locations, demand-forecasting was completed using the popular Microsoft Excel spreadsheet software; however, this field research study shows the demand for blood is nonstationary and cannot be accurately forecasted using Excel, at least not without writing a macro. Daily demand data was obtained from one of the above blood centers for the period of January 2006 through December 2012, and multiple approaches were used to predict blood demand. At the end of the study, the Box–Jenkins methodology was shown to be the optimal choice to forecast demand. It provided accurate demand forecasts for total blood demand (TBD) and all individual blood types with the exception of type A-. The Box–Jenkins methodology was also easy for practitioners to apply, and required only a small investment in software designed to handle this approach.
Source: Operations Research for Health Care - Category: Hospital Management Source Type: research