The status of NCAR's Integrated Turbulence Forecasting Algorithm (ITFA) for predicting upper level clear-air turbulence is reviewed. The essence of the ITFA procedure is to dynamically adjust turbulence diagnostic thresholds and their relative weights in assigning clear-air turbulence forecasts based on recent and past turbulence observations (i.e., pilot reports or PIREPs). This procedure allows the algorithm to minimize forecast errors due to uncertainties in diagnostic performance and thresholds. The algorithm is completely automated, and has been running in a semi-operational mode over past three years, enabling us to derive yearly and seasonal performance statistics based on comparisons to turbulence PIREPs. Using these statistics as a guide, we have been able to intercompare individual diagnostic performance, as well as test various diagnostic thresholding and weighting strategies. The effect of these strategies on performance will be reviewed based on both statistically based performance analyses and case studies.