Taylor Leonard | Georgia Institute of Technology (original) (raw)
Papers by Taylor Leonard
Strategic Decision-Making in Trauma Systems
Lecture Notes in Computer Science, Dec 27, 2023
Proceedings of the 13th International Conference on Data Science, Technology and Applications DATA, 2024
This study aims to establish a quantitative construct for enterprise risk assessment and optimal ... more This study aims to establish a quantitative construct for enterprise risk assessment and optimal portfolio investment to achieve the best aviation security. We first analyze and model various aviation transportation risks and establish their interdependencies via a topological overlap network. Next, a multi-objective portfolio investment model is formulated to optimally allocate security measures. The portfolio risk model determines the best security capabilities and resource allocation under a given budget. The computational framework allows for marginal cost analysis which determines how best to invest any additional resources for the best overall risk protection and return on investment. Our analysis involves cascading and inter-dependency modeling of the multi-tier risk taxonomy and overlaying security measures. The model incorporates three objectives: (1) maximize the risk posture (ability to mitigate risks) in aviation security, (2) minimize the probability of false clears, and (3) maximize the probability of threat detection. This work presents the first comprehensive model that links all resources across the 440 federally funded airports in the United States. We experimented with several computational strategies including Dantzig-Wolfe decomposition, column generation, particle swarm optimization, and a greedy heuristic to solve the resulting intractable instances. Contrasting the current baseline performance to some of the near-optimal solutions obtained by our system, our solutions offer improved risk posture, lower false clear, and higher threat detection across all the airports, indicating a better risk enterprise strategy and decision process under our system. The risk assessment and optimal portfolio investment construct are generalizable and can be readily applied to other risk and security problems.
Journal of Strategic Innovation and Sustainability
We offer a quantitative construct for optimizing security measure investments, to achieve the mos... more We offer a quantitative construct for optimizing security measure investments, to achieve the most costeffective deterrence and detection capabilities for the U.S. Customs and Border Patrol (CBP). We constructed a large-scale multiple-objective portfolio optimization integer program that rapidly returns good Pareto optimal results. The model incorporates the utility of each measure and the probability of success, along with multiple other objectives. To the best of our knowledge, our work presents the first mathematical model that optimizes security strategies for the CBP and is the first to introduce a utility factor to emphasize deterrence and detection impact. The model accommodates different resources, constraints, and various types of objectives. The solution methodologies being put in place are complex, current state-of-the-art, and very effective.
: Past research proposed that it is possible to forecast cargo demand using time series models an... more : Past research proposed that it is possible to forecast cargo demand using time series models and that there exists potential cost savings in the way that Civilian Reserve Air Fleet (CRAF) is used for cargo airlift. United States Transportation Command (USTRANSCOM) performs annual fixed-buys of CRAF to support airlift needs. Forecasted cargo demand would allow for reasonably accurate cargo projections vs. the current expected value estimation. Accurate forecasting allows for greater fixed-buys, further incentivizing CRAF airlines as well as reducing the number of additional aircraft purchases during the quarterly and monthly buys. Multiple forecasting models are constructed and the results compared. A Monte Carlo simulation using a discrete pallet destinations distribution and a discrete pallet port arrival date distribution (based on historical data) outputs a month of projected pallet weights (with date and destination) that are equivalent to the forecasted cargo amount. The simu...
Journal of Strategic Innovation and Sustainability, 2020
We offer a quantitative construct for optimizing security measure investments, to achieve the mos... more We offer a quantitative construct for optimizing security measure investments, to achieve the most costeffective deterrence and detection capabilities for the U.S. Customs and Border Patrol (CBP). We constructed a large-scale multiple-objective portfolio optimization integer program that rapidly returns good Pareto optimal results. The model incorporates the utility of each measure and the probability of success, along with multiple other objectives. To the best of our knowledge, our work presents the first mathematical model that optimizes security strategies for the CBP and is the first to introduce a utility factor to emphasize deterrence and detection impact. The model accommodates different resources, constraints, and various types of objectives. The solution methodologies being put in place are complex, current state-of-the-art, and very effective.
OPERATIONAL PLANNING OF CHANNEL AIRLIFT MISSIONS USING FORECASTED DEMAND, 2013
Past research proposed that it is possible to forecast cargo demand using time series models and ... more Past research proposed that it is possible to forecast cargo demand using time series models and that there exists potential cost savings in the way that Civilian Reserve Air Fleet (CRAF) is used for cargo airlift. United States Transportation Command (USTRANSCOM) performs annual "fixed-buys" of CRAF to support airlift needs. Forecasted cargo demand would allow for reasonably accurate cargo projections vs. the current expected value estimation.
International Journal of Applied Industrial Engineering, 2017
Retailers rely on strong marketing programs to make sales in a competitive environment. One of th... more Retailers rely on strong marketing programs to make sales in a competitive environment. One of the
biggest sectors of marketing is email loyalty programs. The industry uses Click Through Rate to
measure the effectiveness of an email marketing campaign. Click Through Rate measures the ratio
between emails sent to customers and the number of customers that open and click on the email’s
link. The purpose of this analysis is to define the relationship between the rate of emails sent and
the amount of emails opened, opened and clicked, sent to spam, and unsubscribed. The authors’ 800
million records of data consisted of all of the emails sent by a large retail company to its approximate
8 million customers and their responses to those emails for 12 weeks during the winter holiday season.
The results show that the optimal send rate of 6.4 emails per week leads to an increase of emails
opened and clicked while minimizing the number of users that unsubscribe, and, assuming open
source revenue per click figures apply, would generate an additional $1.3 million in weekly revenue.
Strategic Decision-Making in Trauma Systems
Lecture Notes in Computer Science, Dec 27, 2023
Proceedings of the 13th International Conference on Data Science, Technology and Applications DATA, 2024
This study aims to establish a quantitative construct for enterprise risk assessment and optimal ... more This study aims to establish a quantitative construct for enterprise risk assessment and optimal portfolio investment to achieve the best aviation security. We first analyze and model various aviation transportation risks and establish their interdependencies via a topological overlap network. Next, a multi-objective portfolio investment model is formulated to optimally allocate security measures. The portfolio risk model determines the best security capabilities and resource allocation under a given budget. The computational framework allows for marginal cost analysis which determines how best to invest any additional resources for the best overall risk protection and return on investment. Our analysis involves cascading and inter-dependency modeling of the multi-tier risk taxonomy and overlaying security measures. The model incorporates three objectives: (1) maximize the risk posture (ability to mitigate risks) in aviation security, (2) minimize the probability of false clears, and (3) maximize the probability of threat detection. This work presents the first comprehensive model that links all resources across the 440 federally funded airports in the United States. We experimented with several computational strategies including Dantzig-Wolfe decomposition, column generation, particle swarm optimization, and a greedy heuristic to solve the resulting intractable instances. Contrasting the current baseline performance to some of the near-optimal solutions obtained by our system, our solutions offer improved risk posture, lower false clear, and higher threat detection across all the airports, indicating a better risk enterprise strategy and decision process under our system. The risk assessment and optimal portfolio investment construct are generalizable and can be readily applied to other risk and security problems.
Journal of Strategic Innovation and Sustainability
We offer a quantitative construct for optimizing security measure investments, to achieve the mos... more We offer a quantitative construct for optimizing security measure investments, to achieve the most costeffective deterrence and detection capabilities for the U.S. Customs and Border Patrol (CBP). We constructed a large-scale multiple-objective portfolio optimization integer program that rapidly returns good Pareto optimal results. The model incorporates the utility of each measure and the probability of success, along with multiple other objectives. To the best of our knowledge, our work presents the first mathematical model that optimizes security strategies for the CBP and is the first to introduce a utility factor to emphasize deterrence and detection impact. The model accommodates different resources, constraints, and various types of objectives. The solution methodologies being put in place are complex, current state-of-the-art, and very effective.
: Past research proposed that it is possible to forecast cargo demand using time series models an... more : Past research proposed that it is possible to forecast cargo demand using time series models and that there exists potential cost savings in the way that Civilian Reserve Air Fleet (CRAF) is used for cargo airlift. United States Transportation Command (USTRANSCOM) performs annual fixed-buys of CRAF to support airlift needs. Forecasted cargo demand would allow for reasonably accurate cargo projections vs. the current expected value estimation. Accurate forecasting allows for greater fixed-buys, further incentivizing CRAF airlines as well as reducing the number of additional aircraft purchases during the quarterly and monthly buys. Multiple forecasting models are constructed and the results compared. A Monte Carlo simulation using a discrete pallet destinations distribution and a discrete pallet port arrival date distribution (based on historical data) outputs a month of projected pallet weights (with date and destination) that are equivalent to the forecasted cargo amount. The simu...
Journal of Strategic Innovation and Sustainability, 2020
We offer a quantitative construct for optimizing security measure investments, to achieve the mos... more We offer a quantitative construct for optimizing security measure investments, to achieve the most costeffective deterrence and detection capabilities for the U.S. Customs and Border Patrol (CBP). We constructed a large-scale multiple-objective portfolio optimization integer program that rapidly returns good Pareto optimal results. The model incorporates the utility of each measure and the probability of success, along with multiple other objectives. To the best of our knowledge, our work presents the first mathematical model that optimizes security strategies for the CBP and is the first to introduce a utility factor to emphasize deterrence and detection impact. The model accommodates different resources, constraints, and various types of objectives. The solution methodologies being put in place are complex, current state-of-the-art, and very effective.
OPERATIONAL PLANNING OF CHANNEL AIRLIFT MISSIONS USING FORECASTED DEMAND, 2013
Past research proposed that it is possible to forecast cargo demand using time series models and ... more Past research proposed that it is possible to forecast cargo demand using time series models and that there exists potential cost savings in the way that Civilian Reserve Air Fleet (CRAF) is used for cargo airlift. United States Transportation Command (USTRANSCOM) performs annual "fixed-buys" of CRAF to support airlift needs. Forecasted cargo demand would allow for reasonably accurate cargo projections vs. the current expected value estimation.
International Journal of Applied Industrial Engineering, 2017
Retailers rely on strong marketing programs to make sales in a competitive environment. One of th... more Retailers rely on strong marketing programs to make sales in a competitive environment. One of the
biggest sectors of marketing is email loyalty programs. The industry uses Click Through Rate to
measure the effectiveness of an email marketing campaign. Click Through Rate measures the ratio
between emails sent to customers and the number of customers that open and click on the email’s
link. The purpose of this analysis is to define the relationship between the rate of emails sent and
the amount of emails opened, opened and clicked, sent to spam, and unsubscribed. The authors’ 800
million records of data consisted of all of the emails sent by a large retail company to its approximate
8 million customers and their responses to those emails for 12 weeks during the winter holiday season.
The results show that the optimal send rate of 6.4 emails per week leads to an increase of emails
opened and clicked while minimizing the number of users that unsubscribe, and, assuming open
source revenue per click figures apply, would generate an additional $1.3 million in weekly revenue.