Mehmet A. Begen

• Ghandi, F.; Davtalab-Olyaie, M.; Asgharian, M.; Begen, M. A.; Saadatmandi, A., (Forthcoming), "Pareto-optimal peer evaluation in context-dependent DEA", Operational Research
  Abstract: Context-dependent data envelopment analysis (DEA) based on cross-efficiency evaluation has been proposed to present more meaningful measures of attractiveness and progress scores, cross-attractiveness, and cross-progress scores, by considering the distance between the decision-making unit (DMU) under evaluation and the entire evaluation context overall. Using an illustrative example, we show that the state of the art existing method does not guarantee to produce a non-dominated cross- attractiveness (cross-progress) scores vector. This raises concerns about the reliability and universal acceptance of the derived scores. Thus, we investigate the Pareto optimality of these peer evaluation scores and introduce a concept of optimality in line with the dominance notion. We subsequently propose a multi-objective model designed to produce all non-dominated cross-attractiveness (cross-progress) score vectors. We introduce two perspectives – a common base and an individualized approach. The common base method determines the same vector of weights for all units in the evaluation context to produce non-dominated scores. In contrast, the individualized based approach empowers each DMU within the evaluation context to assess the attractiveness (progress) score of DMUs under evaluation at a specific level based on their own distinct criteria. This methodology is aligned with the inherent desire of the DMUs within the evaluation context to have the most substantial impact on the evaluation of cross-attractiveness (cross-progress) for the DMUs under the assessment at a specific level. We illustrate our proposed methods with a real-world examples to yield non-dominated cross-attractiveness (cross-progress) scores.
• Davtalab-Olyaie, M.; Begen, M. A.; Yang, Z.; Asgharian, M., 2024, "Incentivization in Centrally Managed Systems: Inconsistencies Resolution", Omega, December 129
  Abstract: In centrally managed systems (CMSs), the need for incentivization systems at the local management level is crucial to optimize overall performance. Three alternative incentive systems have emerged within the centralized resource allocation (CRA) framework, aiming to measure the contribution of decision-making units (DMUs) in CMSs. However, we identify inconsistencies within these approaches and present them through illustrative examples. First, existing methods may struggle to effectively distinguish between CRA-efficient and CRA-inefficient DMUs, potentially resulting in inappropriate penalties or rewards for some the DMUs. Second, they may encounter difficulty in differentiating among CRA-efficient DMUs, especially when dealing with non-extreme DMUs or masked data within the dataset. Third, these methods may lack precision in measuring the impact of non-extreme CRA-efficient DMUs on overall performance. To address these limitations, we first highlight certain misconceptions related to individual efficiency within CMSs in the existing literature. Subsequently, we establish a fundamental characterization of individual efficient DMUs by outlining necessary and sufficient conditions for categorizing a DMU as CRA-efficient. For the second and third limitations, we adopt an endogenous perspective to quantify the influence of each CRA-efficient DMU. This involves calculating the maxi- mum potential contribution of the DMU under evaluation in constructing the projection points of other DMUs. Furthermore, we propose a new method to handle masked data well in differentiating among CRA-efficient DMUs. We show the validity and applicability of our approaches using a real dataset.

  Link(s) to publication:

  http://dx.doi.org/10.1016/j.omega.2024.103160

• Klein, A.; Begen, M. A., 2024, "COVID-19 Policy Response Analysis: A Canadian Perspective", International Journal of Environmental Research and Public Health, June 21(6)
  Abstract: Abstract The COVID-19 pandemic highlighted the challenges that go into effective policymaking. Facing a public health crisis of epic proportion, government bodies across the world sought to manage the spread of infectious disease and healthcare-system overwhelm in the face of historic economic instability and social unrest. Recognizing that COVID-19 debates and research are still actively ongoing, this paper aims to objectively compare COVID-19 responses from countries across the world that exhibit similar economic and political models to Canada, identify notable failures, successes, and key takeaways to inform future-state pandemic preparedness.

  Link(s) to publication:

  http://dx.doi.org/10.3390/ijerph21060787

• Lofti, A.; Jiang, Z.; Naoum-Sawaya, J.; Begen, M. A., 2024, "Modeling Sales of Multigeneration Technology Products in the Presence of Frequent Repeat Purchases: A Fractional Calculus-Based Approach", Production and Operations Management, May 33(5)
  Abstract: Frequently releasing a new product generation has become a common practice to sustain sales over time, thus accurately forecasting the sales trajectory of each product generation plays a vital role in the short-, medium-, and long-term planning of a firm. Classic multigeneration diffusion models do not incorporate within-generation repeat purchases, making them unusable for product lines with high rates of such purchases. Concentrating on technology products, we develop a multigeneration sales model to fill this void. We demonstrate that the new model can be used for predictive and prescriptive analytics. Our empirical results show that the new model estimates and forecasts sales more accurately than a state-of-the-art benchmark model that does not account for within-generation repeat purchases, underscoring the importance of incorporating repeat purchases. Furthermore, we use two different versions of our model to examine market entry timing under two main strategies, i.e., (i) a phase-out transition strategy in which firms continue to sell the old generation after the release of the new generation, and (ii) a total transition strategy in which firms discontinue the old generation after the introduction of the new generation. Our results indicate that the repeat purchases rate determines whether it is optimal to expedite or delay the new product launch, underscoring the importance of incorporating repeat purchases in market entry strategies.

  Link(s) to publication:

  http://dx.doi.org/10.1177/10591478241240744

• Fišar, M.; Greiner, B.; Huber, C.; Katok, E.; Ozkes, A. I.; Begen, M. A.; Management Science Reproducibility Collaboration, ., 2024, "Reproducibility in Management Science", Management Science, March 70(3): 1343 - 2022.
  Abstract: With the help of more than 700 reviewers, we assess the reproducibility of nearly 500 articles published in the journal Management Science before and after the introduction of a new Data and Code Disclosure policy in 2019. When considering only articles for which data accessibility and hardware and software requirements were not an obstacle for reviewers, the results of more than 95% of articles under the new disclosure policy could be fully or largely computationally reproduced. However, for 29% of articles, at least part of the data set was not accessible to the reviewer. Considering all articles in our sample reduces the share of reproduced articles to 68%. These figures represent a significant increase compared with the period before the introduction of the disclosure policy, where only 12% of articles voluntarily provided replication materials, of which 55% could be (largely) reproduced. Substantial heterogeneity in reproducibility rates across different fields is mainly driven by differences in data set accessibility. Other reasons for unsuccessful reproduction attempts include missing code, unresolvable code errors, weak or missing documentation, and software and hardware requirements and code complexity. Our findings highlight the importance of journal code and data disclosure policies and suggest potential avenues for enhancing their effectiveness.This paper was accepted by David Simchi-Levi, behavioral economics and decision analysis?fast track.Supplemental Material: The online appendices and data are available at https://doi.org/10.1287/mnsc.2023.03556.

  Link(s) to publication:

  https://doi.org/10.1287/mnsc.2023.03556

  http://dx.doi.org/10.1287/mnsc.2023.03556

• Begen, M. A.; Rodrigues, F.; Rice, T.; Zaric, G. S., 2024, "A Forecasting Tool for a Hospital to Plan Inbound Transfers of COVID-19 Patients from Other Regions", Bmc Public Health, February 24(505)
  Abstract: Background: In April 2021, Ontario, Canada, was at the peak of its third wave of the COVID-19 pandemic. Intensive Care Unit (ICU) capacity in the Toronto metropolitan area was insufficient to handle local COVID patients. As a result, some patients from the Toronto metropolitan area were transferred to other regions. Methods: A spreadsheet-based Monte Carlo simulation tool was built to help a large tertiary hospital plan and make informed decisions about the number of transfer patients it could accept from other hospitals. The model was implemented in Microsoft Excel to enable it to be widely distributed and easily used. The model estimates the probability that each ward will be overcapacity and percentiles of utilization daily for a one-week planning horizon. Results: The model was used from May 2021 to February 2022 to support decisions about the ability to accept transfers from other hospitals. The model was also used to ensure adequate inpatient bed capacity and human resources in response to various COVID-related scenarios, such as changes in hospital admission rates, managing the impact of intra-hospital outbreaks and balancing the COVID response with planned hospital activity. Conclusions: Coordination between hospitals was necessary due to the high stress on the health care system. A simple planning tool can help to understand the impact of patient transfers on capacity utilization and improve the confidence of hospital leaders when making transfer decisions. The model was also helpful in investigating other operational scenarios and may be helpful when preparing for future outbreaks or public health emergencies.

  Link(s) to publication:

  http://dx.doi.org/10.1186/s12889-024-18038-3

• Zacharias, C.; Liu, N.; Begen, M. A., 2024, "Dynamic Interday and Intraday Scheduling", Operations research, January 72(1): 317 - 335.
  Abstract: Adaptive Patient Flow Management Appointment scheduling has significant clinical, operational, and economical impact on healthcare systems. An informed scheduling strategy that can effectively match patient demand and service capacity dynamically is vital for the business of medical providers, quality of care, and patient satisfaction. By regulating patient flow via an appointment system, healthcare providers can mitigate arrival process variability and improve operational performance. The simultaneous consideration of appointment day (interday scheduling) and time of day (intraday scheduling) in dynamic scheduling decisions is an important theoretical and practical problem that has remained open because of its stochastic nature, complex structure, and large dimensionality. Zacharias et al. (2022) fill this critical gap in the literature. They introduce a novel dynamic programming framework, designed with the intention of bridging two independently established streams of literature, and to leverage their latest advances in tackling the joint problem. They advance the theory of the field to provide a rigorous and practically implantable solution. The simultaneous consideration of appointment day (interday scheduling) and time of day (intraday scheduling) in dynamic scheduling decisions is a theoretical and practical problem that has remained open. We introduce a novel dynamic programming framework that incorporates jointly these scheduling decisions in two timescales. Our model is designed with the intention of bridging the two streams of literature on interday and intraday scheduling and to leverage their latest theoretical developments in tackling the joint problem. We establish theoretical connections between two recent studies by proving novel theoretical results in discrete convex analysis regarding constrained multimodular function minimization. Grounded on our theory, we develop a practically implementable and computationally tractable scheduling paradigm with performance guarantees. Numerical experiments demonstrate that the optimality gap is less than 1% for practical instances of the problem.

  Link(s) to publication:

  https://pubsonline.informs.org/doi/10.1287/opre.2022.2342

  http://dx.doi.org/10.1287/opre.2022.2342

• Begen, M. A.; Odegaard, F.; Sadeghi, J., 2024, "Intra-provincial benchmark analysis of COVID-19 in Canada", INFOR: Information Systems and Operational Research, January 62(3): 1 - 42.
  Abstract: The COVID-19 pandemic posed unheralded challenges to people, business, government at all levels (federal, provincial, regional), and society at large. In addition to the direct consequences of taking care of infected people, which in some countries led to a virtual collapse of the healthcare system, the pandemic strained eldercare, employment, economic growth, and exacerbated mental health and social problems. During the first year of the pandemic, researchers? and policy makers? main focus was on ?flattening the curve,? and on predictive modeling of infections and deaths. In this paper we present a non-parametric data-driven descriptive analysis based on Data Envelopment Analysis to assess COVID-19 in ten Canadian provinces over the two year period 2020 to 2021. The objective is to derive worst- and best-case intra-provincial benchmarks to assess if and to what extent the situation could have been worse respectively better. To take account for any indirect socio-economic impact our analysis incorporates official monthly unemployment rates and a stringency index reflecting the level of social policy restrictions imposed by the provincial governments. A major contribution of the model framework is that it provides a mechanism for measuring the impact of the two main strategies in curbing the pandemic, namely vaccination and social policy restrictions. As a robustness check, the bench-mark results are compared against bias-corrected efficiency measures.

  Link(s) to publication:

  https://doi.org/10.1080/03155986.2024.2302298

  http://dx.doi.org/10.1080/03155986.2024.2302298

• Begen, M. A.; Odegaard, F.; Sadeghi, J., (Forthcoming), "On aggregation of technical and revenue efficiency measures", Journal of Productivity Analysis: 1 - 16.
  Abstract: We discuss and propose new results regarding combined production efficiencies of an aggregate group, such as industry or region. Based on the production economic notion of aggregate technology, we present an algebraic representation of the aggregation output set, and use this to derive a new decomposition of the aggregation unit’s technical efficiency, e.g. a group’s technical efficiency. Our novel decomposition can identify more clearly the inefficiency due to the underlying units and the inefficiency due to the aggregation. For the special single output case, the proposed aggregation measure is consistent with the extant literature, and identical to Farrell’s measure of structural efficiency of a group. However, for the general multiple outputs case we show that there is a gap between the proposed measure and the extant literature. We illustrate the formal results with a numerical example based on a dataset from literature.

  Link(s) to publication:

  https://doi.org/10.1007/s11123-023-00710-2

  http://dx.doi.org/10.1007/s11123-023-00710-2

• Lyons, J. S. F.; Begen, M. A.; Bell, P. C., 2023, "Surgery Scheduling and Perioperative Care: Smoothing and Visualizing Elective Surgery and Recovery Patient Flow", Analytics, August 2(3): 656 - 675.
  Abstract: This paper addresses the practical problem of scheduling operating room (OR) elective surgeries to minimize the likelihood of surgical delays caused by the unavailability of capacity for patient recovery in a central post-anesthesia care unit (PACU). We segregate patients according to their patterns of flow through a multi-stage perioperative system and use characteristics of surgery type and surgeon booking times to predict time intervals for patient procedures and subsequent recoveries. Working with a hospital in which 50+ procedures are performed in 15+ ORs most weekdays, we develop a constraint programming (CP) model that takes the hospital’s elective surgery pre-schedule as input and produces a recommended alternate schedule designed to minimize the expected peak number of patients in the PACU over the course of the day. Our model was developed from the hospital’s data and evaluated through its application to daily schedules during a testing period. Schedules generated by our model indicated the potential to reduce the peak PACU load substantially, 20-30% during most days in our study period, or alternatively reduce average patient flow time by up to 15% given the same PACU peak load. We also developed tools for schedule visualization that can be used to aid management both before and after surgery day; plan PACU resources; propose critical schedule changes; identify the timing, location, and root causes of delay; and to discern the differences in surgical specialty case mixes and their potential impacts on the system. This work is especially timely given high surgical wait times in Ontario which even got worse due to the COVID-19 pandemic.

  Link(s) to publication:

  http://dx.doi.org/10.3390/analytics2030036

• Akioyamen, P.; Begen, M. A., 2023, "A Spatio-Temporal Analysis of OECD Member Countries’ Health Care Systems: Effects of Data Missingness and Geographically and Temporally Weighted Regression on Inference", International Journal of Environmental Research and Public Health, June 20(13): 6265 - 6265.
  Abstract: Determinants of health care quality and efficiency are of importance to researchers, policy-makers, and public health officials as they allow for improved human capital and resource allocation as well as long-term fiscal planning. Statistical analyses used to understand determinants have neglected to explicitly discuss how missing data are handled, and consequently, previous research has been limited in inferential capability. We study OECD health care data and highlight the importance of transparency in the assumptions grounding the treatment of data missingness. Attention is drawn to the variation in ordinary least squares coefficient estimates and performance resulting from different imputation methods, and how this variation can undermine statistical inference. We also suggest that parametric regression models used previously are limited and potentially ill-suited for analysis of OECD data due to the inability to deal with both spatial and temporal autocorrelation. We propose the use of an alternative method in geographically and temporally weighted regression. A spatio-temporal analysis of health care system efficiency and quality of care across OECD member countries is performed using four proxy variables. Through a forward selection procedure, medical imaging equipment in a country is identified as a key determinant of quality of care and health outcomes, while government and compulsory health insurance expenditure per capita is identified as a key determinant of health care system efficiency.

  Link(s) to publication:

  http://dx.doi.org/10.3390/ijerph20136265

• Dogan, S.; Okuyan, H. M.; Bal, T.; Çabalak, M.; Begen, M. A., 2023, "Relationship of thrombospondin-1 and thrombospondin-2 with hematological, biochemical and inflammatory markers in COVID-19 patients", Turkish Journal of Biochemistry-Turk Biyokimya Dergisi, June 48(4): 368 - 375.
  Abstract: Objectives: Roles of thrombospondin-1 (TSP-1) and thrombospondin-2 (TSP-2) in tissue repair and inflammation are well-documented, but the association of their serum expressions with the pathogenesis of COVID-19 remains unclear. We investigate the roles of TSP-1 and TSP-2 in COVID-19. Methods: 106 SARS-CoV-2 infected patients and 23 healthy people were enrolled in our study. COVID-19 patients were divided into two groups as non-severe and severe. TSP-1 and TSP-2 concentrations were measured with an enzyme-linked Immunosorbent Assay, and blood markers were analyzed with routine laboratory techniques. Results: COVID-19 patients had significantly higher TSP-1 and TSP-2 levels than healthy controls. TSP-1 and TSP-2 positively correlated with inflammatory markers, including ESR, CRP, PCT, ferritin, and biochemical parameters such as ALT, AST, BUN, CK, and LDH. In addition, TSP-1 and TSP-2 were negatively correlated with hematological markers such as LYM, EOS, and HGB. Receiver operating characteristic analyses revealed that COVID-19 may be predicted with TSP-1 levels over 189.94 ng/mL and TSP-2 levels higher than 0.70 ng/mL. Conclusions: Our analysis suggests that TSP-1 and TSP-2 expressions at the systemic level may have clinical importance for COVID-19.

  Link(s) to publication:

  https://doi.org/10.1515/tjb-2022-0265

  http://dx.doi.org/10.1515/tjb-2022-0265

• Sadeghi, J.; Begen, M. A.; Odegaard, F., 2023, "Refined bounds for the non-Archimedean ϵ in DEA", Computers & Operations Research, June 154: 106163 - 106163.
  Abstract: Motivated by measuring the Pareto–Koopmans efficiency of public service units, we present refined bounds for the crucial non-Archimedean infinitesimal, aka epsilon. In non-parametric efficiency modeling, epsilon plays a key role as a multiplication factor to the sum of input and output slacks in the objective function. However, selecting an appropriate value for epsilon is non-trivial. It has to be sufficiently small to guarantee the envelopment model is bounded (or multiplier model to be feasible), yet large enough to provide managerial insight and not cause computational problems. Furthermore, the appropriate value is context dependent on the input and output metrics, and sensitive to assumptions regarding constant or variable returns-to-scale. Finally, different epsilon values may lead to drastically different ordering of the relative efficiency measures. To guarantee consistent relative order of the evaluated units we provide two refined bounds. The first, positive efficiency measures, serves as a precursor to ensure the obtained Pareto–Koopmans efficiency measures are positive and well-defined. The second and main contribution, robust efficiency measures, ensures the relative efficiency measures are provably consistent. We illustrate our bounds and their implications from an evaluation study of twelve public healthcare centers.

  Link(s) to publication:

  https://www.sciencedirect.com/science/article/pii/S0305054823000278

  http://dx.doi.org/10.1016/j.cor.2023.106163

• Naderi, B.; Begen, M. A.; Zaric, G. S.; Roshanaei, V., 2023, "A Novel and Efficient Exact Technique for Integrated Staffing, Assignment, Routing, and Scheduling of Home Care Services Under Uncertainty", Omega-International Journal of Management Science, April 116: 102805 - 102805.
  Abstract: We model and solve an integrated multi-period staffing, assignment, routing, and scheduling for home care services under uncertainty. The goal is to construct a weekly schedule that adheres to related operational considerations and determines optimal staffing of caregivers by minimizing caregivers’ fixed- and overtime costs. For tractability, we incorporate a priori generated visit patterns—an existing practical approach that deals effectively with hard assignment decisions in. First, we propose a novel mixed-integer program (MIP) for the nominal problem. We then incorporate uncertainty in service and travel times and develop a robust counterpart by hybridizing interval and polyhedral uncertainty sets. Second, we show that there is a special mathematical structure within the model that allows us to develop a novel logic-based Benders branching-decomposition algorithm that systematically delays the resolution of difficult routing/ scheduling problems and efficiently solves both the nominal and robust MIP models. Using a dataset from the literature, we show that CPLEX can solve our nominal and robust models with an average optimality gaps of 44.56% and 45.53%, respectively. Using the same dataset, we demonstrate that our new exact technique can solve our nominal and robust mixed-integer models to an average optimality gap of 2.8% and 4.5%, respectively. Third, we provide practical insights into (i) the price of robustness and (ii) the impacts of nurse flexibility and overtime. The average total cost does not increase beyond 12.7% than the nominal solution and the cost-savings of nurse flexibility is approximately five times higher than that of overtime.

  Link(s) to publication:

  https://www.sciencedirect.com/science/article/pii/S0305048322002110

  http://dx.doi.org/10.1016/j.omega.2022.102805

• BAYAZIT, B.; UÇARKUS, G.; ÇAGLAR GENÇOSMAN, B.; Begen, M. A., 2022, "Meta-Heuristic Algorithms based on Integer Programming for Shelf Space Allocation Problems", The European Journal of Science and Technology, November (41): 100 - 117.
  Abstract: Retail shelf space management, which is one of the most complex aspects of retailing, can be defined as determining when, where and in what quantities products will be displayed and dynamically updating the display considering changing market conditions. Although it is an important problem, research papers that study rectangular arrangement of products to optimize profit are limited. In this paper, we determine rectangular facing units of products to maximize profit for shelf space allocation and the display problem. To solve our two-dimensional shelf space allocation problem, we develop two matheuristic algorithms by using integer programming and genetic algorithm (TP-GA) and integer programming and firefly algorithm (TP-ABA) meta-heuristics together. The performances of the mathheuristics were compared with a real-world dataset from a bookstore. TP-GA and TP-ABA methods were able to generate near optimal solutions with an average of 4.47% and 4.57% GAPs, respectively. We can also solve instances up to 900 products. These matheuristic algorithms, which are successful in the two-dimensional shelf assignment problem, can also be used to solve similar problems such as allocation of books in a bookstore, allocation of product families in a grocery store, or display of advertisements on websites.

  Link(s) to publication:

  https://dergipark.org.tr/tr/pub/ejosat/issue/72892/1121006

  http://dx.doi.org/10.31590/ejosat.1121006

For more publications please see our Research Database