The 2024-25 Academic Year in Research: Part 3

The 2024-25 academic year saw a number of successes on the research front. Jumping to November of 2024, Turner College management professor Johnny Ho published a study in the International Journal of Services and Standards demystifies the potential of generative AI by evaluating the performance of ChatGPT in addressing operations management problems and questions. The findings of his study reveal that ChatGPT generally performs well but struggles with aspects of Bloom’s taxonomy, suggesting its limitations in higher-order cognitive tasks. This particular finding underscores the importance of collaboration among educators, learners, and generative AI to enhance educational outcomes. Lastly, the study also explores the role of prompt engineering and custom GPTs in improving education and learning in operations management courses. Taken as a whole, Ho's study provides significant insights into operations management education and pedagogy, unveiling many opportunities for future research.

     TSYS School computer scientist Riduan Abid was also active in November of 2024, publishing a study in IEEE Access examines a promising approach for running deep learning models at the edge that employs split neural networks. Split neural networks feature a neural network architecture with multiple early exit points, allowing the model to make confident decisions at earlier layers without processing the entire network. This not only reduces memory and computational demands but it also makes split neural networks well-suited for edge computing applications. As the use of split neural networks expands, ensuring their safety — particularly their robustness to perturbations — becomes crucial for deployment in safety-critical scenarios. Abid's research presents the first in-depth study on the robustness of split Edge Cloud neural networks. In doing so it reviews state-of-the-art robustness certification techniques and evaluate split neural networks robustness using the auto_LiRPA and Auto Attack libraries, comparing them to standard neural networks. The results demonstrate that split neural networks reduce average inference time by 75‘% and certify four to 10 times more images as robust, while improving overall robustness accuracy by 1% to 10%.

     Next, a study by Turner College management professor Kevin Hurt appearing in Conflict Resolution Quarterly develops a theoretical model and evidence-based propositions depicting the interrelationships between servant leadership, organizational trust, and affective conflict. The paper positions affective conflict, which is a type of disagreement that occurs when people focus on their personal feelings and relationships with others, rather than the task at hand, as a negative moderating influence between servant leadership and organizational trust in order to present a solution to mitigate the negative effects of affective conflict. The study first reviews the relationship between servant leadership and organizational trust and develop propositions linking the constructs. Hurt then discusses the impact of affective conflict on the servant leadership—organizational trust relationship and develops additional propositions linking those constructs. Finally, the study presents the Walk-In-The-Woods process as a method to moderate the impact of affective conflict on the servant leadership—organizational trust relationship and to efficiently resolve conflict and preserve the positive effects of servant leadership. As Hurt explains, the Walk-In-The-Woods is a structured conflict resolution process that is based on the concept of principled negotiation. Its purpose is to expand the range of interests that are involved in a conflict resolution setting to reach an efficient and amicable solution.

     Turner College economist Frank Mixon adds a fourth November 2024 study that examines whether compensating wages and price differentials exist across cities in the U.S. The study, published by the Journal of Regional Analysis & Policy, calculates the net implicit monetary values of the cost of living in cities with a higher cost of entertainment offerings. These net implicit monetary values reflect an individual’s willingness to pay for, or willingness to accept, life in cities with a higher cost of entertainment offerings. Viewed in this way, a city’s entertainment offerings are not clearly an amenity or disamenity. The paper explores the idea that the implicit monetary value of households’ preferences with regard to employment when wages do not fully reflect the cost entertainment offerings. Using data from the 2018 American Community Survey for cities in North Carolina, results from a seemingly unrelated regression approach suggest that although the cost of entertainment offerings is capitalized into both wages and rents, individuals receive compensating differentials (i.e., larger increases in wages than in rents) for living in cities with high costs associated with entertainment offerings.

     December of 2024 began with an interdisciplinary contribution from Turner College management professor Kirk Heriot and TSYS School computer scientist Rania Hodhod. Their study moves beyond anecdotal evidence to uncover deeper insights into entrepreneurial decision-making by employing artificial intelligence (AI) and natural language processing (NLP) in order to analyze feedback from 100 entrepreneurs. By processing large datasets, the study, which appears in Electronics, identifies patterns and correlations that provide valuable perspectives on the evolving landscape of entrepreneurship, with implications for education, policy, and practice.

     Sliding window aggregation, which extracts summaries from data streams, is a core operation in streaming analysis. However, real-world data streams often involve out-of-order data and exhibit burst data characteristics, which pose performance challenges to these sliding window algorithms. To address this challenging issue, a study in IEEE Transactions on Knowledge and Data Engineering by TSYS School computer scientist Yi Zhou proposes Gecko - a novel sliding window aggregation algorithm that supports bulk cache eviction. Gecko works by leveraging a granular-based eviction strategy for various bulk sizes, enabling efficient bulk eviction while maintaining the performance close to that of in-order stream algorithms for single evictions. For large data bulks, Gecko performs coarse-grained eviction at the chunk level, followed by fine-grained eviction using leftward binary tree aggregation (LTA) as a complementary method. Moreover, Gecko partitions data based on chunks to prevent the impacts of out-of-order data on other chunks, thereby enabling efficient handling of out-of-order data streams. Extensive experiments conducted by Zhou to evaluate the performance of Gecko demonstrate its superior performance over other solutions. In real-world data scenarios, Gecko performs 1.7 to 3.5 times better than the state-of-the-art algorithm while also demonstrating the best latency performance among all compared schemes.

     Closing out December of 2024 is a study in the Journal of Financial Planning by Turner College professor of finance Gisung Moon that delves into the implications of incorporating Social Security as an asset in investment portfolios alongside stocks and bonds. As Moon asserts, by treating Social Security as a bond-like asset, portfolio allocations are influenced, leading to higher initial investments in stocks that gradually decrease as retirees grow older. The study underscores the significance of factoring in Social Security when making retirement investment decisions, as it can impact risk and return profiles. According to Moon, financial planners can glean valuable insights from this research to better guide clients in retirement planning, considering variables such as gender, marital status, life expectancy, and risk tolerance. Lastly, the paper recommends that additional research along similar lines be conducted in order to deepen our understanding of how these factors shape efficient portfolios that include Social Security.