Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WB50

3 - Optimization Based Method for Seismic Rehabilitation of Water Pipe Networks Azam Boskabadi, PhD Candidate, University of Texas at Arlington, Arlington, TX, United States, Jay Michael Rosenberger, Mohsen Shahandashti, Binaya Pudasaini We introduce a stochastic programming model for water pipe rehabilitation problem with a recourse flow function to maximize the output flow in the subgraph after earthquake. The stochastic programming model starts with an initial rehabilitation policy in water pipes which shows the decision of rehabilitating water pipes before earthquake, then it considers several random scenarios for the broken pipes after earthquake. We use Benders Decomposition in our study, the master problem finds the best X or rehabilitation policy before the earthquake that maximizes the expected SSI, while each recourse problem Shirish Rao, University of Texas-Arlington, Arlington, TX, 76019, United States, Victoria C. P. Chen, Jay Michael Rosenberger, Atefe Makhmalbaf, Takuma Mitani Building structures have significant impact on environment and energy consumption. Buildings can be designed so that their energy consumption is reduced by using energy efficient materials. The “cradle-to-grave” cycle is a criterion for the environmental effects of materials and minimizing this effect along with energy consumption is of great interest. The goal of the research is to compare different experimental designs and statistical modeling methods to help inform our approach for a multivariate, multi-response green building framework. The results of the experiments are studied by using Treed regression and Multi Adaptive Regression Splines techniques to find influential factors. n WB49 North Bldg 230 Energy Policy and Planning II Contributed Session Chair: Erin Mayfield, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States 1 - Corporate Responses to Public Pressures and Price during an Electric Power Interruption Shigeharu Okajima, Associate Professor, Osaka University of Economics, 2-2-8 Osumi, Higashiyodogawa-ku, Osaka, 533-8533, Japan This study investigates corporate electricity consumption behaviors during a power interruption in a natural field experiment. It is found that in general, firms reduce electricity consumption in response to public pressures, but not electricity price increases, and that increasing electricity price is yet effective to reduce electricity consumption of such firms that do not react to public pressures. 2 - Finding Multiple Equilibria in a Dynamic Energy Bilateral Contract Scheduling Javier Contreras, Universidad Castilla La Mancha, Ciudad Real, Spain, Reinaldo Crispiniano Garcia, Matheus Barbosa, Barbara Macedo, Daniel Monteiro In the deregulated electricity markets the involved parties, the buyers and the sellers, aim to deliver electricity in time intervals during the contract period to obtain the highest possible profits. This paper implements a model where the parties by a compromise approach achieve a possible equilibria including the Nash and the Raiffa-Kalai-Smorodinski (RKS) ones. An approach in prices and quantities is applied transforming the nonlinear and nonconvex set of the inequalities into a mixed integer linear problem (MILP). A procedure to find all Nash and RKS equilibria is developed by generating “holes added as linear constraints to the feasibility region showing the feasibility of the approach. 3 - A Robust Contingency Constrained Unit Commitment with an N- AK Security Criterion Jaehee Jeong, Korea Advanced Institute of Science & Technology, Dept. of Industrial and Systems Eng,, KAIST, 291 Daehak-Ro, Yuseong-Gu, Daejeon, 305-701, Korea, Republic of, Sungsoo Park In the power grid systems, the N-k contingency-constrained unit commitment (CCUC) model has been adopted to guarantee the system’s reliability. One of the drawbacks of N-k CCUC model is that its robust solution is too conservative. To mitigate this drawback, we propose N-ak CCUC model which excludes some extreme cases with low probability. An economic and reliable robust solution can be achieved by a proper criterion setting. Our model is solved by a Benders’ decomposition type algorithm with maximum feasible subsystem cuts. Numerical results on a modified IEEE 118-bus system are reported. assigns maximum output flow in the created subgraph. 4 - Sustainability Assessment for Green Building

4 - Recent Advances in MILP Formulations for the Unit Commitment Problem Bernard Knueven, Sandia National Laboratories, P.O. Box 5800, MS 1326, Albuquerque, NM, 87185-1326, United States, Jim Ostrowski, Jean-Paul Watson This paper presents recent work on MILP formulations for the unit commitment problem (UC). UC is that of deciding which power generators to schedule to meet anticipated energy demand, and is of critical importance in the operation of power systems. We review existing MILP formulations along with novel formulations made by combining various ideas from the literature and present a comprehensive computational study of said formulations. In addition, we provide reference implementations of the reviewed formulations in the Pyomo modeling language, along with a library of UC test instances. 5 - A Multiobjective Model of Environmental and Labor Market Impacts of Natural Gas Development Decisions in Appalachia Erin Mayfield, Carnegie Mellon University, Pittsburgh, PA, United States, Jared L. Cohon, Allen Robinson With the advancement of drilling and fracturing technology and the resulting rapid acceleration of natural gas production, there have been corresponding human health, climate, ecosystem, and labor market impacts. We focus our analysis on Appalachia, a region with a long history of both resource extraction and poverty, as well as, the current location of the highest producing natural gas reservoir in the US. A multiobjective optimization model is formulated in which we optimize spatially-explicit environmental and labor market impacts with respect to decisions regarding the magnitude and timing of natural gas activity. We also incorporate distributive and temporal equity constraints. Decision Making Contributed Session Chair: Gary R. Waissi, Arizona State University, Cave Creek, AZ, 85331, United States 1 - Using Social Media to Advocate and Enable Knowledge Sharing Within Agile Teams Deepa Umakanth, Arizona State University, Backus Mall, Mesa, AZ, 85128, United States, Gary R. Waissi (adviser) This paper discusses a case study of using social media to advocate and enable knowledge sharing within agile teams. Knowledge sharing involves sharing of knowledge with employees, within teams and organizations. Social media aide in the sharing and exchange of knowledge. With a good implementation of Agile framework, tools and online processes to share knowledge effectively, a company can grow effectively and will be able to face the competition. The results indicate that knowledge sharing using online tools is easy, fast and is the current trend with Agile frameworks. 2 - Finding a High Quality Initial Solution for the Traveling Salesman Problem (TSP) Pragya Kaushal, Arizona State University, Tempe, AZ, United States, Gary Waissi We propose an efficient algorithm to preprocess the fully dense symmetric network to find a good, feasible, solution for the TSP. The algorithm generates a sequence of spanning trees, and using those constructs a “layered networkö. This layered network will include a feasible solution for the TSP. This solution can serve as the starting solution, for example, for the TABU search. We hypothesize, that the quality of the initial solution provided by the proposed algorithm will improve the performance of the TABU search in terms of number of iterations required to reach a high level TSP solution. 3 - A Novel Matrix Technique for Finding an Efficient Solution for the Traveling Salesman Problem (TSP) Gary R. Waissi, Arizona State University, Cave Creek, AZ, 85331, United States We propose a polynomial algorithm to process the fully dense symmetric network to find a good, feasible, solution for the TSP. The algorithm processes the cost matrix representing the network to generate a feasible TSP tour. One iteration includes making an assignment in a selected row or column, then omitting that row or column from further consideration, declaring the column or row opposite to the first assignment as “tabuö, and thereby reducing the problem size. The matrix reduction and “tabu assignment assure that the process will result into a complete tour. n WB50 North Bldg 231A

459