Publications – J. Haden Boone

- Military Logistics Planning for Expeditionary Warfare
  with Matthew Ford, Samuel Tan, Mathieu Dahan, Peter Frazier, Yongjia Song, and Huseyin Topaloglu. Technical Paper
  Preprint / ResearchGate / Show AbstractHide Abstract
  
  Abstract
  Problem definition: In times of geopolitical instability, the United States Marine Corps (USMC) may be required to quickly deploy expeditionary forces to contested areas. This article develops a decision-support tool that designs military logistics plans for supporting flow and sustainment of such forces in expeditionary environments. Methodology: We propose a scheduled service network design model to optimize the routing and scheduling of various transportation vehicles (connectors) and commodities across an intermodal expeditionary logistics network. Formulated as a mixed-integer program (MIP), the objective of this model is to minimize penalties incurred for late force closure and fulfillment of demand, while accounting for connector capacities, speeds, and ranges, as well as the availability of commodities. Results and managerial implications: We illustrate the application of our model through a fictitious warfare scenario set in Southern California designed to reflect USMC training scenarios. Our results highlight the effectiveness of the proposed MIP in creating complex intermodal logistics plans to ensure timely fulfillment of expeditionary demands. This framework offers the first optimization-based decision-support tool for expeditionary logistics planning, contrasting with the current qualitative analyses conducted by military logistics planners. Our findings provide the USMC with valuable granular and aggregated insights into the efficiency and vulnerability of their logistics network.

- Inspection Game with Imperfect Detection Technology
  with Mathieu Dahan. Submitted.
  Preprint / ResearchGate / Show AbstractHide Abstract
  
  Abstract
  In this article, we analyze a network inspection game in which two players, an attacker and a defender, coordinate multiple resources across a discrete set of locations. Specifically, the attacker chooses multiple components within a network to target, while the defender chooses a subset of disjoint collections of components to position detectors. However, the detection technology utilized is assumed to be fallible: efficacy of a detector is assumed to attenuate across the components of the network with respect to some metric (e.g. distance) based on technology utilized and network properties, resulting in successful detection probabilities being both location-and component-specific. The defender (respectively, attacker) aims to minimize (respectively, maximize) the expected number of attacks on the network that go undetected. We solve this large-scale zero-sum game by a) computing an analytical solution for a mixed-strategy Nash (or saddle-point) equilibrium, and b) provide intuition behind player strategies and decisions in equilibrium by highlighting components and quantities of importance to the players, introducing concepts of saturation and critical detection performance. We show how these results can be leveraged to influence decisions on what types of detection technology to purchase.

- SQSTS: A Sequential Procedure for Estimating Steady-State Quantiles Using Standardized Time Series
  with Athanasios Lolos, Christos Alexopoulos, David Goldsman, Kemal Dinçer Dingeç, Anup Mokashi, and James R. Wilson. Accepted to Journal of Simulation.
  Preprint / GaTech / Show AbstractHide Abstract
  
  Abstract
  We develop and evaluate SQSTS, an automated sequential procedure for computing point estimators and confidence intervals (CIs) for steady-state quantiles based on the simulation analysis methods of Standardized Time Series (STS) and sectioning as the latter method is applied to batch quantile estimators and the full-sample quantile estimator. The variance parameter associated with the full-sample quantile estimator is estimated by a combination of variance-parameter estimators that are based on the two aforementioned methods of simulation analysis and are asymptotically independent as the batch size increases with a fixed number of batches. Building on the theoretical framework of Alexopoulos et al.(2022) for STS-based quantile estimation, we establish a multivariate central limit theorem for increasing batch sizes that is the basis for the key steps in SQSTS for controlling the growth of the batch size on successive iterations of the procedure. Extensive experimentation with a variety of test processes revealed that SQSTS performed well compared to its competitors in terms of the CI’s estimated coverage probability; and it outperformed those competitors with regard to average sample-size requirements.

- Infrastructure Inspection with Imperfect Detection Technology
  with Mathieu Dahan. American Control Conference (ACC), May 2023.
  Conference / DOI / Show AbstractHide Abstract
  
  Abstract
  We consider a two-player zero-sum inspection game, in which a limited number of detectors are coordinated in an infrastructure system according to a probability distribution to detect multiple attacks from a strategic opponent. Detection is assumed to be imperfect and depends on the detectors’ technology and the infrastructure’s properties. We analytically characterize Nash equilibria of this large-scale game for problem instances where each component is detected from one detector location and the attacker has limited resources. Our equilibrium analysis provides a new criticality assessment of the infrastructure’s components in strategic settings. It also demonstrates new equilibrium behavior from the players that intricately depends on their amount of resources, as well as the detection technology and infrastructure topology.

- A Sequential Method for Estimating Steady-State Quantiles Using Standardized Time Series
  with Athanasios Lolos, Christos Alexopoulos, David Goldsman, Kemal Dinçer Dingeç, Anup Mokashi, and James R. Wilson. Winter Simulation Conference (WSC), December 2022.
  Conference / DOI / Show AbstractHide Abstract
  
  Abstract
  We propose SQSTS, an automated sequential procedure for computing confidence intervals (CIs) for steady-state quantiles based on Standardized Time Series (STS) processes computed from sample quantiles. We estimate the variance parameter associated with a given quantile estimator using the order statistics of the full sample and a combination of variance-parameter estimators based on the theoretical framework developed by Alexopoulos et al. in 2022. SQSTS is structurally less complicated than its main competitors, the Sequest and Sequem methods developed by Alexopoulos et al. in 2019 and 2017. Preliminary experimentation with the customer delay process prior to service in a congested M/M/1 queueing system revealed that SQSTS performed favorably compared with Sequest and Sequem in terms of the estimated CI coverage probability, and it significantly outperformed the latter methods with regard to average sample-size requirements.

- Steady-State Quantile Estimation Using Standardized Time Series
  with Athanasios Lolos, Christos Alexopoulos, David Goldsman, Kemal Dinçer Dingeç, and James R. Wilson. Winter Simulation Conference (WSC), December 2020.
  Conference / DOI / Show AbstractHide Abstract
  
  Abstract
  Extending developments of Calvin and Nakayama in 2013 and Alexopoulos et al. in 2019, we formulate point and confidence-interval (CI) estimators for given quantiles of a steady-state simulation output process based on the method of standardized time series (STS). Under mild, empirically verifiable conditions, including a geometric-moment contraction (GMC) condition and a functional central limit theorem for an associated indicator process, we establish basic asymptotic properties of the STS quantile-estimation process. The GMC condition has also been proved for many widely used time-series models and a few queueing processes such as M/M/1 waiting times. We derive STS estimators for the associated variance parameter that are computed from nonoverlapping batches of outputs, and we combine those estimators to build asymptotically valid CIs. Simulated experimentation shows that our STS-based CI estimators have the potential to compare favorably with their conventional counterparts computed from nonoverlapping batches.