Cell formation (CF) is the most well-known problem in designing cellular manufacturing systems. In the generalised version of the CF problem, each part has a number of alternative process plans. Two issues should be addressed in the generalised cell formation (GCF) problem: selecting the best process plan for each part and assigning machines to the cells. In addition to the minimisation of movement costs, the selection of the most reliable process plans is an important task for the GCF problem. Although the expected value (EV) method is a straightforward procedure to model stochastic situations, this method has some serious drawbacks. Therefore, we use chance-constrained programming (CCP) to model this problem. To demonstrate the advantages of the proposed model, the model is compared with the traditional GCF model and the EV model through some numerical examples. The results reveal that the proposed model is superior to the other alternatives.