470506 Optimal Integrated Water Management and Shale Gas Supply Chain Planning Under Uncertainty

Monday, November 14, 2016: 2:05 PM
Carmel II (Hotel Nikko San Francisco)
Omar J. Guerra1, Andrés Joaquín Calderon Vergara2, Lazaros G. Papageorgiou3 and Gintaras V. Reklaitis1, (1)School of Chemical Engineering, Purdue University, West Lafayette, IN, (2)Centre for Process Systems Engineering, UCL (University College London), London, United Kingdom, (3)Centre for Process Systems Engineering, Centre for Process Systems Engineering, University College London, London, United Kingdom

The production of unconventional fossil fuels, i.e. shale gas, has gained increasing importance in the energy sector. For instance, under the New Policy Scenario (the International Energy Agency (IEA) base line scenario), the IEA estimated that shale gas production will increase by almost two-fold (from 331 billion cubic meters (bcm) to 941 bcm) from 2013 to 2040 [1]. However, the development of shale gas resources imposes important challenges and potential risks to water resources. Specifically, the depletion, potential contamination, and degradation of both underground and surface water sources have been raised as impediments to the wider development of shale gas resources [2, 3]. Additionally, the design and planning of both water management and shale gas supply chains involve uncertainties that are inherent to the system: these include shale play productivity and wastewater composition, as well as uncertainties associated with market conditions, i.e. gas demand and prices.

This work deals with the development and implementation of a two-stage stochastic optimization approach for the design and planning of the integrated water management and shale gas supply chain. First, a global sensitivity analysis is carried out using the framework developed by the authors [4, 5] to assess and rank uncertain parameters in the integrated supply chain. Then, a Monte Carlo sampling technique is combined with Sobol’s sensitivity indices [6, 7] to compute the effect of uncertainties on the net present value (NPV) performance metric used in the aforementioned framework. Based on the outcomes of the sensitivity analysis, a two-stage stochastic model involving a Mixed Integer Linear Program (MILP) is developed. The first-stage decisions in this model consist of the investment in drilling and fracturing operations as well as in transportation and processing facilities for both water and shale gas. The second-stage decisions are associated with operational issues related to water management and gas delivery. The potential benefits of modeling uncertainty and implementing stochastic models are quantified through two metrics: expected value of perfect information (EVP) and value of stochastic solution (VSS) [8]. Additionally, scenario reduction approaches are evaluated to mitigate computational challenges.


[1] IEA. “World Energy Outlook” 2015. doi:10.1787/weo-2014-en.

[2] Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, Abad JD. “Impact of shale gas development on regional water quality”. Science 2013;340:1235009. doi:10.1126/science.1235009.

[3] Olmstead SM, Muehlenbachs LA, Shih J, Chu Z, Krupnick AJ. “Shale gas development impacts on surface water quality in Pennsylvania”. Proceedings of the National Academy of Sciences of the United States of America 2013;110:4962–7. doi:10.1073/pnas.1213871110.

[4] Calderón AJ, Guerra OJ, Papageorgiou LG, Siirola JJ, Reklaitis G V. “Preliminary Evaluation of Shale Gas Reservoirs: Appraisal of Different Well-Pad Designs via Performance Metrics”. Industrial & Engineering Chemistry Research 2015;54:10334–49. doi:10.1021/acs.iecr.5b01590.

[5] Guerra OJ, Calderón AJ, Papageorgiou LG, Siirola JJ, Reklaitis G V. “An Optimization Framework for the Integration of Water Management and Shale Gas Supply Chain Design”. Computers & Chemical Engineering 2016;(in press, doi: 10.1016/j.compchemeng.2016.03.025).

[6] I. M. Sobol, “Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates,” Mathematics and Computers in Simulation, vol. 55, pp. 271–280, 2001.

[7] A. Saltelli, S. Tarantola, and K. P.-S. Chan, “A Quantitative Model-Independent Method for Global Sensitivity Analysis of Model Output,” Technometrics, vol. 41, no. May, pp. 39–56, 1999.

[8] J. Birge and F. Louveaux, Introduction to stochastic programming. 2011.

Extended Abstract: File Not Uploaded
See more of this Session: Supply Chain Logistics and Optimization
See more of this Group/Topical: Computing and Systems Technology Division