Insights for Critical Alarm-Based Warning Systems from a Risk Analysis of Commercial Aviation Passenger Screening

Please scroll down for article-it is on subsequent pages With 12,500 members from nearly 90 countries, INFORMS is the largest international association of operations research (O.R.) and analytics professionals and students. INFORMS provides unique networking and learning opportunities for individual professionals, and organizations of all types and sizes, to better understand and use O.R. and analytics tools and methods to transform strategic visions and achieve better outcomes. For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org

Section: Applications and Case Studiesmentioning

confidence: 99%

From the Editors: Decision Analysis Focus and Trends

Bier

French

2020

Cutoff Threshold Decisions for Classification Algorithms with Risk Aversion

Hupman

2022

Classification algorithms predict the class membership of an unknown record. Methods such as logistic regression or the naïve Bayes algorithm produce a score related to the likelihood that a record belongs to a particular class. A cutoff threshold is then defined to delineate the prediction of one class over another. This paper derives analytic results for the selection of an optimal cutoff threshold for a classification algorithm that is used to inform a two-action decision in the cases of risk aversion and risk neutrality. The results provide insight to how the optimal cutoff thresholds relate to the associated costs and the sensitivity and specificity of the algorithm for both the risk neutral and risk averse decision makers. The optimal risk averse threshold is not reliably above or below the optimal risk neutral threshold, but the relation depends on the parameters of a particular application. The results further show the risk averse optimal threshold is insensitive to the size of the data set or the magnitude of the costs, but instead is sensitive to the proportion of positive records in the data and the ratio of costs. Numeric examples and sensitivity analysis derive further insight. Results show the percent value gap from a misspecified risk attitude increases as the specificity of the classification algorithm decreases.

Partitioning the Expected Value of Countermeasures with an Application to Terrorism

John,

Dillon,

Burns

et al. 2024

Benefit–cost analyses are critical to support U.S. agencies’ programmatic decision making. These analyses are particularly challenging when one of the benefits is adversary deterrence. This paper presents a framework for calculating the value of deterrence related to countermeasures implemented to mitigate an attack by an adaptive adversary. We offer an approach for partitioning the benefit of countermeasures into three components: (1) threat reduction (deterrence), (2) vulnerability reduction, and (3) consequence mitigation. The benefit of a countermeasure is measured by the expected value of countermeasure implementation (EVCI) attributable to a specific countermeasure. It is based on the concept of expected value of imperfect control, defined as the difference in the expected values of alternatives with and without countermeasures. The EVCI represents all the benefits of implementing the countermeasure and is derived from three sources: (1) changes in attack probability (threat reduction from deterrence), (2) changes in detection probability (vulnerability reduction), and (3) changes in the distribution of attack outcomes (consequence mitigation). We partition the EVCI and estimate the portion attributable to each of these three sources to quantify the unique benefit of each. We provide two applications of the partitioning methodology using examples from the published literature that examine countermeasures designed to protect commercial aircraft against man-portable air defense systems. The proposed framework provides an approach for explicitly accounting separately for deterrence, vulnerability reduction, and consequence mitigation in benefit–cost analyses. It provides quantifiable insights into how countermeasures reduce terrorism risk. Funding: This material is based upon work supported by the U.S. Department of Homeland Security under [Grant Award 22STESE00001-02-00]. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security. This award was made to Northeastern University and the University of Southern California is a sub-awardee. This work was also supported by the National Science Foundation [Grant 2027296] awarded to Decision Research.