Stochastic programming approach to optimal design and operations of shale gas supply chain under uncertainty

In this paper, we propose the first stochastic model addressing the optimal design and operations of the comprehensive shale gas supply chain, where uncertainties of estimated ultimate recovery (EUR) in each shale well are considered. The resulting mixed-integer linear programming (MILP) model covers the well-to-wire life cycle of shale gas, which consists of a number of stages including freshwater acquisition, shale well drilling, fracking, and completion, shale gas production, wastewater management, shale gas processing, electricity generation as well as transportation and storage. In order to reduce the model size and the number of scenarios, we use a sample average approximation approach to generate scenarios based on the EUR distribution derived from actual historical data. To demonstrate the proposed stochastic model and solution approach, we present a case study based on Marcellus shale play to maximize the total expected profit of this shale gas supply chain network.