In this work, a mathematical programming model for optimal design of large scale packing/production systems in chemical and consumer goods manufacturing is presented. The model is capable of handling all possible deterministic sequence dependent changeover times and packing/production times as well as being sufficiently general to cope with uncertainties introduced by the probabilistic behaviour of operations, resource breakdowns, repair times and changeovers when such uncertainties can be defined mathematically. These uncertainties are often common in industry. This model is formulated in terms of random variables for packing/production times, packing/production volumes and the number of occurrence of downtimes. It is shown that the probabilistic model can be reduced to a mixed integer linear program (MILP), once the probability distribution functions for the uncertainties are defined. The model presented here, utilises an efficient ’Continuous Time Formulation’ in which the timings of the activity events are mathematically defined relative to one another in a logical setting. The resulting model is solved to obtain a robust design which minimises an objective such as the variance of the production times of the packing/production systems. The advantage of this formulation is that it is possible to use a standard solver to solve large scale problems which makes the approach very attractive to industry.