A general modular methodology for the simultaneous optimization of the supply-chain network and the production systems of a general industrial gas producer is developed and implemented in a C++ program. The formulation and solution algorithm are specifically designed to be able to work on-line and to determine the optimal assignments of production site output to customer demand in the supplychain and the corresponding optimal operating conditions for the production plants in integrated fashion. Here, the production network is not simply modelled as a set of product sources, rather the model is detailed enough to allow effective and feasible optimization of the entire system. Moreover, the proposed approach can be easily combined with the rolling horizon technique to mitigate the uncertainties in demand. The modelling strategies, employed for the supply-chain network and the production sites, along with the solution approach, adopted for the resulting optimization problem, are detailed. BzzMath library classes are used to meet the computational efficiency requirements for on-line applications. The effectiveness of the proposed methodology is demonstrated on a case study involving a portion of the real supply-chain and network of production facilities of Linde Gas Italia S.r.l. (C) 2015 Elsevier Ltd. All rights reserved.