In diesel engines the optimization of engine-out emissions, combustion noise and fuel consumption requires the experimental investigation of the effects of different injection strategies as well as of a large number of engine operating variables, such as scheduling of pilot and after pulses, rail pressure, EGR rate and swirl level. Due to the high number of testing conditions involved full factorial approaches are not viable, whereas Design of Experiment techniques have demonstrated to be a valid methodology. However, the results obtained with such techniques require a subsequent critical analysis, so as to investigate the cause and effect relationships between the set of engine operating variables and the combustion process characteristics that affect pollutant formation, noise of combustion and engine efficiency.To this purpose, the zero-dimensional multizone diagnostic combustion model developed at ICEAL was applied for the combustion and emission formation analysis in two different diesel engines, for various sets of injection strategies and engine operating parameters. The experimental data were acquired at the highly dynamic test rig of ICEAL, both in a EURO V low compression ratio diesel engine with a twin-stage turbocharger, equipped with piezoelectric injectors, and in a PCCI low compression ratio diesel engine equipped with solenoid injectors. The model results were discussed and reported in the well-known ᵩ-T diagrams, which give a synthetic representation of the local thermodynamic charge conditions during the mixture formation and premixed-diffusion combustion processes. The rail pressure increase was found to be an effective means to improve fuel-charge premixing and to lower the average local equivalence ratio of the charge during premixed combustion, so leading to a decrease in soot formation. As regards the effects of cooled high-pressure EGR on combustion, it was shown that an increase of its rate does not significantly affect the average equivalence ratio during premixed combustion, which is associated to the soot formation rate. Finally, a proper calibration of the dwell-time between pilot and main injection, so as to have the main injection pulse center-of-gravity phased in correspondence to the start of pilot burning, resulted to produce a reduction of CO and soot emissions higher than 50% with respect to baseline case, without any deterioration of NOx emissions.