Young massive clusters (YMCs) are recently formed astronomical objects with unusually high star formation rates. We propose the collision of giant molecular clouds (GMCs) as one of the likely formation mechanisms of YMCs, consistent with the YMC conveyor-belt formation mode concluded by other authors. We performed smoothed particle hydrodynamical (SPH) simulations of colliding clouds, and we explored the parameter space such as the collision speed, initial cloud density, turbulence, and stellar feedback to form YMCs via this mechanism. On collision speed, initial cloud density, and turbulence, we found that in general, greater collision speed, greater initial cloud density, and lower turbulence returns greater star formation rate induced by the collision. We found that colliding clouds with relative velocity & 25 km/s, initial cloud density & 250 cubic-cm, and turbulence ≈ 2.5 km/s allow for the formation of clusters that resemble the real YMCs observed in the Milky Way. On stellar feedback, we first developed a prescription to create stars that are sampled from a realistic initial mass function (IMF) from groups of sink particles. Stars that follow an IMF are required so that feedback and cluster dynamics can be modelled more accurately. Nonetheless, in parsec-scale simulations, IMF is usually not well sampled because sink particles are usually not resolved as individual stars, at the same time each sink is not large enough to sample a whole population of stars. The method developed, i.e. the grouped star formation prescription aims to solve this problem by first grouping the sink particles according to their positions, velocities, and ages, then the group masses are used to sample the IMF and form star particles. This method is tested and is shown to be robust in simulations of different physical scales, from sub-parsec to kilo-parsec, and from static isolated clouds to colliding clouds. With suitable grouping parameters, the grouped star formation method can form stars that can follow the IMF and mimic the original stellar distributions and velocity dispersion. With realistic star particles, we proceed to our investigation of the effect of feedback on YMC formation via cloud-cloud collisions. We developed a code to compute the photoionisation feedback in star-forming region. Using the initial condition of one of our previous colliding clouds, we included photoionisation feedback in our simulations. In general, we found that the feedback model creates less massive but larger clusters in size. Nonetheless, in our colliding cloud setup, clusters that resemble real YMCs can still form regardless of feedback. We note that the effect of feedback on cluster formation is dependent on the environment as noted by other authors