2016 INFORMS Annual Meeting Program

TA28

INFORMS Nashville – 2016

TA29 202A-MCC Managing Capacity in Energy Markets Through Demand and Supply-side Interventions Sponsored: Manufacturing & Service Oper Mgmt, Sustainable Operations Sponsored Session Chair: Charles J Corbett, University of California - Los Angeles, Los Angeles, CA, United States, charles.corbett@anderson.ucla.edu 1 - Energy Efficiency Contracting In Supply Chains Under Asymmetric Bargaining Power Ali Shantia, HEC Paris, 1 rue de la Liberation, Jouy-en-Josas, 78350, France, ali.shantia@hec.edu, Sam Aflaki, Andrea Masini Evidence shows that suppliers refrain from investing in energy efficiency (EE) measures because they fear that a buyer with greater bargaining power will use the EE-related cost reductions to push prices down, in the purchase bargaining process, and thereby further reduce the supplier’s profit margin. In a supply chain consisting of a buyer and a supplier, this study analyses the effect of relative bargaining power and technology uncertainty on the supplier’s decision to invest in energy efficiency measures. We analyse price commitment and shared investment contracts as potential coordination mechanisms and compare them in their ability to boost EE investment by the supplier. 2 - An Analysis Of Time-based Pricing In Electricity Supply Chains Asligul Serasu Duran, Kellogg School of Management, 2001 Sheridan Road, 5th floor, Evanston, IL, 60208, United States, a-duran@kellogg.northwestern.edu, Baris Ata, Ozge Islegen This study builds a framework for the retail electricity market to empirically evaluate the impact of time-based tariffs on the electricity supply chain. We find that optimal time-based tariffs reduce peak demand, but do not change consumers’ electricity bills significantly. Time-of-use tariffs with predetermined rates can capture most of the benefits of real-time prices. The environmental impact of time-based tariffs depends on the characteristics of the electricity market under study. 3 - Investments In Renewable And Conventional Energy: The Role Of Operational Flexibility Kevin Shang, Duke University, Durham, NC, United States, khshang@duke.edu, Gurhan Kok, Safak Yucel We study capacity investments of a utility firm in renewable and conventional energy sources with different levels of operational flexibility, i.e., the ability to quickly ramp up or down the output of a generator. We consider supply characteristics of conventional and renewable sources and derive the optimal capacity investment portfolio. We find that inflexible sources (e.g., nuclear energy) and renewables are substitutes; flexible sources (e.g., natural gas) and renewables are complements. 4 - Explaining The Variation In Progress In The Us Nuclear Industry Christian Blanco, University of California - Los Angeles, Los Angeles, CA, United States, cblanco@anderson.ucla.edu, Felipe Caro, Charles J Corbett We examine the factors that influenced the US nuclear power production efficiency and safety over time. TA30 202B-MCC Studies in Service Operations Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Robert Batt, Wisconsin School of Business, UW - Madison, Madison, WI, United States, rbatt@bus.wisc.edu 1 - Heart Failure transitions: Staffing Follow-up Clinics To Reduce Readmissions Itai Gurvich, Kellogg School of Management, i- gurvich@kellogg.northwestern.edu, Benjamin Grant, Jan A Van Mieghem, Kannan Mutharasan Heart failure (HF) readmissions are a major driver of cost and health care utilization. Timely follow-up of patients post-discharge represents an evidence- based intervention proven to reduce readmission rates. Patients discharged after HF hospitalization are scheduled to meet a cardiologist in the outpatient clinic. Meeting targets for timely follow-up requires appropriate capacity planning for these clinics that takes into account the inpatient-discharge variability. An intervention based on simple safety capacity rules and more aggressive utilization of existing capacity resulted in more than doubling the fraction of patients seen within one week of discharge.

3 - Selling To Socially Connected Customers Ruslan Momot, INSEAD, Fontainebleau, France, ruslan.momot@insead.edu, Elena Belavina, Karan Girotra We study the value of different kinds of social network information and illustrate its use. We build a model of a social network of strategically interacting customers who value exclusive ownership of the product and are heterogeneous in the number of friends (degree) and proclivity for social comparisons (conspicuity). We find that high-conspicuity customers within intermediate-degree segments are the firm’s best targets. Our analysis reveals how they should be selectively targeted by the firms with information on either (or both) of the customer traits. We find that information about degree is more valuable than information about conspicuity and that the two are substitutes. 4 - Subscription Box Business Models: Pricing And Quality Decisions Basak Kalkanci, Georgia Institute of Technology, basak.kalkanci@scheller.gatech.edu, Necati Tereyagoglu We model the value of online subscription box business model for a consumer who chooses the replenishment frequency (or timing) of a frequently used durable good. We explore the seller’s pricing and quality decisions under the online subscription box business model, and evaluate the performance of such a model in comparison to selling through an offline retail channel. TA28 201B-MCC Sequential Sampling and Optimization Sponsored: Applied Probability Sponsored Session Chair: Raghu Pasupathy, Purdue University, West Lafayette, IN, United States, pasupath@purdue.edu 1 - Sequential Stopping Rules For Simulation Problems Where Variance Estimation Is Difficult Jing Dong, Northwestern University, jing.dong@northwestern.edu, Peter W Glynn We solve the sequential stopping problem for a class of simulation problems in which variance estimation is difficult. In particular, we establish the asymptotic validity of sequential stopping procedures for estimators constructed using various cancellation methods. We characterize the limiting distribution of the estimators at stopping times as the error size (the absolute error or the relative error) goes to 0, which is different from the limiting distribution of the estimator constructed based on a fixed size of samples as the sample size goes to infinity. 2 - Probabilistic Bisection Converges Almost As Quickly As Stochastic Approximation Shane Henderson, Professor, Cornell University, 230 Rhodes Hall, Ithaca, NY, 14853, United States, sgh9@cornell.edu, Peter Frazier, Rolf Waeber The probabilistic bisection algorithm (PBA) can be applied to stochastic root finding problems in one dimension. The PBA successively updates a Bayesian prior on the location of the root after using a power-one test at the median of the posterior to estimate the direction of the root from the median. The power-one test has a variable sample size. The PBA has features that we believe make it attractive relative to stochastic approximation for such problems. I will discuss the algorithm and sketch a proof that it converges at a rate arbitrarily close to the canonical “square root” rate of stochastic approximation. 3 - Fixed-Step, Line Search, And Trust-Region Adaptive Sampling Recursions for Simulation Optimization Raghu Pasupathy, Purdue University, pasupath@purdue.edu We present a sequential sampling framework for recursively solving stochastic optimization problems. The framework consists of embedding a globally conver- gent numerical optimization search routine, e.g., line search, trust region, with Monte Carlo sampled estimators of the objective function and gradient. Global convergence to a stationary point depends crucially on a result characterizing the sample size at each iteration. We will outline the conditions that guarantee the attainment of the Monte Carlo canonical rate.

242