A problem of probabilistic hesitant fuzzy multi-attribute decision making with completely unknown probability information is studied. Firstly, aiming at the case that the probabilities of probabilistic hesitant fuzzy elements (PHFEs) are completely unknown, a nonlinear programming model is proposed to determine the unknown probabilities of PHFEs. Secondly, according to the evaluation of decision-makers, the weight of attribute can be determined by entropy weight method. In addition, an approach based on LINMAP method is put forward to determine the probabilistic hesitant fuzzy ideal solutions objectively in the probabilistic hesitant fuzzy environment. Then, the compromise ratio method (CRM), whose basic principle is that the optimal alternative should have the nearest distance from positive ideal solution and the longest distance from negative ideal solution simultaneously, is utilized to rank the alternatives. And more objective and reasonable decision results can be obtained through CRM. Finally, a numerical example is given to illustrate the effectiveness and feasibility of the proposed method.