RECENTLY, INCREASING ATTENTION HAS FOCUSED ON THE ECONOMIC DESIGN OF QUALITY CONTROL CHARTS. E.S. PAGE, WHO DEVELOPED ONE OF THE PIONEERING WORKS IN THIS AREA IN 1953, DISCUSSED THE CHOICE OF SAMPLE SIZE AND CONTROL LIMITS FOR CONTROLLING THE MEAN OF A UNIVARIATE NORMAL POPULATION USING THE SCHEME OF MINIMIZING THE AVERAGE RUN LENGTH OF AN OUT-OF-CONTROL PROCESS. THIS METHOD DEFINES THE AVERAGE RUN LENGTH AS THE AVERAGE NUMBER OF ARTICLES INSPECTED BETWEEN TWO SUCCESSIVE OCCASIONS WHEN RECTIFYING ACTION IS TAKEN. IN TODAY'S COMPLEX SOCIETY, HOWEVER, THE QUALITY OF EACH ITEM DEPENDS ON SEVERAL QUALITY CHARACTERISTICS. STILL, MULTIVARIATE QUALITY CONTROL PROCEDURES HAVE NOT BEEN IMPLEMENTED, PRIMARILY FOR TWO REASONS: (1) A PROCEDURE FOR THE SELECTION OF SAMPLE SIZE HAS NOT BEEN DEVELOPED, AND (2) A PROCEDURE FOR DETERMINING THE VALUE FOR THE CONTROL CHART CONSTANT HAS NOT BEEN FORMULATED. SUCH PROCEDURES ARE PRESENTED IN THIS STUDY. THEY RESULT FROM EXTENDING PAGE'S SCHEME OF MINIMIZING THE AVERAGE RUN LENGTH OF AN OUT-OF-CONTROL PROCESS IN THE CASE OF AN ITEM WITH TWO QUALITY CHARACTERISTICS. RESULTS CAN BE APPLIED TO ITEMS WITH MORE THAN TWO CHARACTERISTICS. IT WAS FOUND THAT FOR CORRELATED QUALITY CHARACTERISTICS, THE SAMPLE SIZE NEEDED TO DETECT SHIFTS IN THE LEVELS OF THESE CHARACTERISTICS DOES NOT ALWAYS DECREASE AS THE MAGNITUDES OF THE SHIFTS IMPORTANT TO DETECT INCREASES. A SMALLER SAMPLE SIZE IS ALSO NEEDED FOR TWO QUALITY CHARACTERISTICS THAN FOR ONE QUALITY ITEMS. TABULAR DATA, MATHEMATICAL EQUATIONS, AND REFERENCES ARE PROVIDED. (RCB)