A computer planning model for blood platelet production and distribution.

We consider a class of policies for stocking hospital blood banks with units of random donor platelet concentrate ('Platelets') based upon scheduled daily deliveries from a regional blood center to replenish the platelet inventory to a fixed 'base stock' level. The measures of interest are the 'shortage rate' (the proportion of days for which the on-hand inventory at the hospital blood bank is insufficient to meet the demand) and the 'outdate rate' (the proportion of total units shipped which are not transfused within the usable life span of 5 days). Our principal results give a predictive model which relates the base stock level to the shortage rate and outdate rate. Our model uses only the mean daily demand as a parameter. It provides a basis to unify the results from other studies which have demonstrated improvements in platelet inventory management in particular hospitals and blood centers.

Blood distribution systems and the exchange of information between hospital blood banks and regional blood centers.

The advantages and disadvantages of centralized or decentralized reactive or predictive blood distribution systems are presented. A decentralized, predictive distribution system, being used by the Long Island Blood Service Division of the Greater New York Blood Program is described and its applicability to most regional blood supply organizations is emphasized.

An evaluation of the CODABAR symbol in blood-banking automation.

The CODABAR system has been used in blood banking automation since 1976 when it was introduced by 16 blood centers and transfusion services throughout the United States in a test program coordinated by the American Blood Commission Committee for Commonality in Blood Banking Automation. Since then CODABAR has been used in sample identification on Groupamatic and Technicon automated blood-typing equipment, in the labeling of blood products and in their distribution. It is widely used by a number of blood centers in the United States and has been adopted by the 14 regional blood centers in the United Kingdom, as well as blood centers in Switzerland, the Netherlands and Japan. Some properties of the CODABAR symbol that are relevant to the high accuracy requirements of blood banking are described and preliminary test data are presented.

A regional blood management system with prescheduled deliveries.

The prescheduling of blood deliveries from regional blood centers can improve blood availability at all transfusion services while reducing outdating, delivery costs and inventories maintained at region blood centers. A procedure for determining optimal distribution policies that maintain the highest possible blood availability at transfusion services while holding the achievable outdating level to a small percentage of the total usage is described. The manner in which this plan is being implemented in Long Island, New York, is presented, as well as a discussion of some preliminary results.

Regional blood center automation: computer surveillance of donor blood processing.

Computer assistance in blood processing is necessary in large blood donor centers if errors are to be eliminated. Sample identification, recording of data, and use of the computer data file in final labeling are most important.

Regional blood center automation.

The operation of an automated regional blood center is outlined. The basic approaches used to achieve economical automation that is compatible with existing manual protocols are described. Finally, the role of centralized, automated blood typing is discussed in the context of the operations of The Greater New York Blood Program which involves a central Groupamatic blood typing facility serving four blood centers.

Setting inventory levels for hospital blood banks.

A method is devised to predict the hospital blood bank inventory of each of the eight blood types that will be adequate on all but a given fraction of days (shortage rate). Analysis of the model confirms and extends prior statistical models and provides a rationale for a number of observations made on some common blood inventory practices. Comparison of the relationship between inventory levels and shortage rates predicted by the model and that calculated from data from 21 hospital blood banks of varying sizes, located in four separate parts of the country, reveals agreement within approximately 10 per cent. A procedure is presented and allows blood banks to select inventory levels and shortage rates based on their mean demand for blood.