PurposeA model study of laser heating process including phase change and molten flow in the melt pool gives physical insight into the process and provides useful information on the influence of melting parameters. In addition, the predictions reduce the experimental cost and minimize the experimental time. Consequently, investigation into laser control melting of the titanium alloy becomes essential. The purpose of this paper is to do this.Design/methodology/approachLaser repetitive pulse heating of titanium surface is investigated and temperature field as well as Marangoni flow in the melt pool is predicted using finite volume approach. The influence of laser scanning speed and laser pulse parameter (defining the laser pulse intensity distribution at the workpiece surface) on temperature distribution and melt size is examined. The experiment is carried out to validate temperature predictions for two consecutive laser pulses.FindingsThe influence of laser scanning speed is significant on the melt pool geometry, which is more pronounced for the laser pulse parameter β=0. Temperature predictions agree with the thermocouple data obtained from the experiment.Research limitations/implicationsAlthough temperature dependent properties are used in the simulations, isotropy in properties may limit the simulations. The laser canning speed is limited to 0.3 m/s, which is good for surface treatment process, but it may slow for annealing treatments.Practical implicationsThe results are very useful to capture insight into the melting process. In addition, the influence of laser scanning speed and laser pulse intensity distribution on the melt formation in the surface vicinity is well presented, which will be useful for those working on laser surface treatment process.Originality/valueThe work is original and findings are new, which demonstrate the influence of laser parameters on the melt pool formation and resulting Marangoni flow.