The method previously used in the Toxicology Laboratories of King Faisal Specialist Hospital and Research Center for determining the zinc concentration in serum by Zeeman atomic absorption spectrometer was improved by modifying the matrix modifier and by changing the heated graphite furnace atomization (HGA) program. After trying several methods we failed to achieve the required precision and the accuracy of methods for serum zinc determination. We changed the matrix modifier to a fifty percent mixture (v/v) of 3.90 grams per liter of ammonium phosphate in Type 1 water with 0.2% nitric acid and 1.0 gram per liter of magnesium nitrate in acidic water (0.2% HNO3) with 0.1% triton X-100 was used as matrix modifier. A twenty-five fold dilution of the sample in matrix modifier was injected on the L'vov's platform of the furnace. In order to reduce the high sensitivity of Zn the furnace program was modified. The method is found very robust. The average reproducibility between inter-runs and intra-run is less than 1.59% with a high degree of accuracy. We used two levels of controls i.e. normal or low level and abnormal or high level. The linearity and the detection limit of the assay were 0.9992 and 0.010 micromol/L respectively. Average recovery of the analyte was 98.65%. The X-Bar and R charts were constructed by using Shewhart's statistical analysis technique to assess the test methodology. It was found that the assay is capable and stable for routine clinical and research analysis. The capability index (C(P)) of the assay, an indicator of the precision, was calculated.