Zero-inflated Poisson regression is  the model for count data with the excess zeros. It supposes that with probability p the only possible observation is 0 and with the probability 1 p a random variable with the Poisson distribution is observed. For instance, when manufacturing equipment is properly aligned, defects might be almost impossible. But when it is misaligned, defects might happen according to a Poisson distribution. Both probability p of the perfect zero defect state and the mean number of defects λ in the imperfect state might depend on covariates. The parameters in this type of models can be estimated using maximum likelihood estimation.