RESUMO
To maximize the availability of platelet concentrates (PC) and minimize their outdate, competing criteria, a computer simulation model of platelet production and distribution was developed. Based on 2 years of actual platelet orders placed with a regional blood center, the simulation program generated daily platelet orders, and also calculated mean demand and standard deviation of demand for each day of the week. The number of PC to be produced on a given day was calculated from: PC to be produced = (mean demand for that day of the week) + Tx (standard deviation of demand for that day of the week)-(PC in inventory on the given day), where T is a selected multiple of the standard deviation. As T increases, availability is maximized, but outdating is expected to increase. Conversely, lower T is associated with less availability, but lower outdate. At the simulated platelet demand level (about 735 PC per week), with 3-day platelet storage life, 99% availability is associated with 1% outdate and distribution of PC of less than 1 day average age. 100% availability, with no outdate is predicted for a 5-day storage life, with no further improvement in logistics with 7-day storage life, at this level of demand. Although logistics of platelet production and distribution vary from center to center, the simulation analysis is generally applicable, and a formal plan has the great advantage of predictive, rather than reactive, platelet production.
