Article ID: | iaor19982247 |
Country: | United States |
Volume: | 3 |
Issue: | 3 |
Start Page Number: | 139 |
End Page Number: | 155 |
Publication Date: | Dec 1997 |
Journal: | International Journal of Operations and Quantitative Management |
Authors: | Balakrishnan Nagraj, Miller Janis L. |
Keywords: | energy |
Natural Gas Pipeline Company (NGPL) stores gas in several underground formations of different size during summer and pumps it to the Chicago area for consumption during winter. The gas absorbs moisture during the storage period and must therefore be dried before it re-enters the pipeline. The waste-water collected during the drying process is temporarily stored in small tanks at each formation, and then transported to a central facility for proper disposal. In order to maintain an uninterrupted gas flow from a formation, the water in its tank must be picked up before the tank becomes full. In this paper, we present a mathematical formulation for the water pickup problem, and describe two separate heuristic approaches for its solution. We then discuss the computational performance of these heuristics using actual gas and water production data from NGPL. The results indicate that the heuristics would be valuable decision making tools for NGPL in the future. Although this paper focuses on the specific problem faced by NGPL, the heuristics are general enough to be applicable to similar routing and scheduling problems faced by other firms.