Louis Luangkesorn

A hybrid modeling approach was proposed and developed as a tool for the economic analysis of poultry breeds used in complex long-term egg production systems. The factors considered included both the stored and collected internal operational data that related to the within-breed historical life-cycle reliability and the related economic data that influenced egg prices and poultry life spans in a poultry farm. In the designed simulation models, the forecasted egg sales prices from a designed… Show more

A hybrid modeling approach was proposed and developed as a tool for the economic analysis of poultry breeds used in complex long-term egg production systems. The factors considered included both the stored and collected internal operational data that related to the within-breed historical life-cycle reliability and the related economic data that influenced egg prices and poultry life spans in a poultry farm. In the designed simulation models, the forecasted egg sales prices from a designed machine learning algorithm were incorporated to evaluate the economic analysis of specific poultry breeds. Our analysis results demonstrated that simulations could be combined with machine learning to serve as a powerful large-scale data analysis tool for the poultry breeds used in complex long-term egg production systems. doi: 10.1109/WSC57314.2022.10015442. Show less

This research describes a pilot program for propagating active learning within engineering education starting with a group of nine interested instructors from two departments. The first and second authors served as the discipline-based coaches for these instructors, and the propagation program involved community discussions, one-on-one coaching, classroom observation, assessment of student perspectives, and feedback to and follow-up with the instructors. This approach aligned with the… Show more

This research describes a pilot program for propagating active learning within engineering education starting with a group of nine interested instructors from two departments. The first and second authors served as the discipline-based coaches for these instructors, and the propagation program involved community discussions, one-on-one coaching, classroom observation, assessment of student perspectives, and feedback to and follow-up with the instructors. This approach aligned with the professional development and coaching literature as well as emergent change strategies identified by
Henderson and colleagues. This work is important because STEM education has not generally taken a research-based approach to dissemination of pedagogical innovations, and research on sustained change is only in its early stages. Using a case study approach involving instructor interviews, documentary data (i.e., discussion notes), and classroom observation, the program was assessed based upon instructor participation and accomplishments (including scholarship of teaching and learning activities), plans for continued active-learning use, and valuation. Of the nine initial instructors,
seven participated in the one-year program until the end, including three who also engaged in scholarship of teaching and learning. All seven used active learning, as confirmed by observation or interview. Based on their interviews, instructors identified the program’s ‘‘people’’ focus, in particular one-on-one coaching and community discussions, as strengths of the program, as supported by the coaching literature. A finding of this research is that benefits were achieved despite non-ideal levels of instructor participation in all program aspects. The goal is to share an implementation and assessment approach with other educators considering relationship-driven, emergent strategies for adoption or expansion of active learning. Show less

This study explored the application of machine learning to generate metamodel approximations of a physics-based fire hazard model. The motivation to generate accurate and efficient metamodels is to improve modeling realism in probabilistic safety assessments where computational burden has prevented broader application of high fidelity models. The process involved scenario definition, generating training data by iteratively running the fire hazard model called CFAST over a range of input space… Show more

This study explored the application of machine learning to generate metamodel approximations of a physics-based fire hazard model. The motivation to generate accurate and efficient metamodels is to improve modeling realism in probabilistic safety assessments where computational burden has prevented broader application of high fidelity models. The process involved scenario definition, generating training data by iteratively running the fire hazard model called CFAST over a range of input space using the RAVEN software, exploratory data analysis and feature selection, an initial testing of a broad set of metamodel methods, and finally metamodel selection and tuning using the R software.

Twenty-five metamodel methods ranging in class and complexity were investigated. Linear models struggled because the physics of fire are non-linear. A k-nearest neighbor (kNN) model fit the vast majority of calculations within ±10% for maximum upper layer temperature and its timing.

The resulting kNN model was compared to an algebraic model typically used in fire probabilistic safety assessments. This comparison illustrated the potential of metamodels to improve modeling realism over simpler models selected for computational feasibility. While the kNN metamodel is a simplification of the higher fidelity model, the error introduced is quantifiable and can be explicitly considered. Show less

Availability of extremely large repositories of healthcare data and new analytic tools have revolutionized efficient access to comprehensive data across large numbers of patients. Advances in tools and methods for analyzing large stores of data enable detection of subtle patterns in these data, even with missing or less than optimal data quality. This chapter introduces approaches to data science, analytics, and knowledge building from health data, including details on a process known as… Show more

Availability of extremely large repositories of healthcare data and new analytic tools have revolutionized efficient access to comprehensive data across large numbers of patients. Advances in tools and methods for analyzing large stores of data enable detection of subtle patterns in these data, even with missing or less than optimal data quality. This chapter introduces approaches to data science, analytics, and knowledge building from health data, including details on a process known as knowledge discovery and data mining (KDDM). Data science or “big data” concepts are introduced, and the fundamentals of data analytics are discussed. The chapter explores the application of data analytic processes in healthcare and describes organizational needs for data science such as personnel and data governance. The chapter concludes with recent advances and challenges in the field and a discussion of future directions in data science, data analytics, and KDDM. Show less

Developing prediction bounds for surgery duration is difficult due to the large number of distinct procedures. The variety of procedures at a multi-speciality surgery suite means that even with several years of historical data a large fraction of surgical cases will have little or no historical data for use in predicting case duration. Bayesian methods can be used to combine historical data with expert judgement to provide estimates to overcome this, but eliciting expert opinion for a… Show more

Developing prediction bounds for surgery duration is difficult due to the large number of distinct procedures. The variety of procedures at a multi-speciality surgery suite means that even with several years of historical data a large fraction of surgical cases will have little or no historical data for use in predicting case duration. Bayesian methods can be used to combine historical data with expert judgement to provide estimates to overcome this, but eliciting expert opinion for a probability distribution can be difficult. We combine expert judgement, expert classification of procedures by complexity category and historical data in a Markov Chain Monte Carlo (MCMC) model and test it against one year of actual surgery cases at a multi-speciality surgical suite. Show less

Organizations often have to make production capacity decisions in a setting where production disruptions occur. This presents managers with a combination of strategic capacity decisions and operational inventory management options to manage disruptions. This paper uses Bayesian methods to analyze operational data where data on parameters required in logistics models are unavailable, then models production and inventory management in systems that have the potential for major disruptions. The… Show more

Organizations often have to make production capacity decisions in a setting where production disruptions occur. This presents managers with a combination of strategic capacity decisions and operational inventory management options to manage disruptions. This paper uses Bayesian methods to analyze operational data where data on parameters required in logistics models are unavailable, then models production and inventory management in systems that have the potential for major disruptions. The produce-to-stock with production disruptions model is applicable to systems where the decision on production rate is coupled with the setting of the base stock level when production disruptions are possible. The model is applied to a proposed food processing facility at a correctional institution that is subject to disruptions due to safety and security issues. Show less

During the 2009 H1N1 pandemic, computational agent-based models (ABMs) were extensively used to evaluate interventions to control the spread of emerging pathogens. However, evaluating different possible combinations of interventions using ABMs can be computationally very expensive and time-consuming. Therefore, most policy studies have examined the impact of a single policy decision.
OBJECTIVE: To apply a sequential experimental design method with an ABM to analyze policy alternatives… Show more

During the 2009 H1N1 pandemic, computational agent-based models (ABMs) were extensively used to evaluate interventions to control the spread of emerging pathogens. However, evaluating different possible combinations of interventions using ABMs can be computationally very expensive and time-consuming. Therefore, most policy studies have examined the impact of a single policy decision.
OBJECTIVE: To apply a sequential experimental design method with an ABM to analyze policy alternatives composed of a combination of school closure and vaccination policies to provide a set of promising "optimal" combinations of policies to control an H1N1-type epidemic to policy makers.
METHODS: We used an open-source agent-based modeling system, FRED (A Framework for Reconstructing Epidemiological Dynamic), to simulate the spread of an H1N1 epidemic in Alleghany County, Pennsylvania, with a census-based synthetic population. We used an approach called best subset selection method to evaluate 72 alternative policies consisting of a combination of options for school closure threshold, closure duration, Advisory Committee on Immunization Practices prioritization, and second-dose vaccination prioritization policies. Using the attack rate as a performance measure, best subset selection enabled us to eliminate inferior alternatives and identify a small group of alternative policies that could be further evaluated on the basis of other criteria.
RESULTS: Our sequential design approach to evaluate a combination of alternative mitigation policies leads to a savings in computational effort by a factor of 2 when examining combinations of school closure and vaccination policies. Best subset selection demonstrates a substantial reduction in the computational burden of a large-scale ABM in evaluating several alternative policies. Our method also provides policy makers with a set of promising policy combinations for further evaluation based on implementation considerations or other criteria. Show less

Emergency managers have to develop plans for responding to disasters within their jurisdiction. This includes coordinating multiple independent agencies participating in the response. While much of this is currently done by use of intuition and expert judgment, models can be used to test assumptions and examine the impact of policies and resource levels. The autonomous nature of responders as well as the rapidly changing information during a disaster suggests that agent based models can be… Show more

Emergency managers have to develop plans for responding to disasters within their jurisdiction. This includes coordinating multiple independent agencies participating in the response. While much of this is currently done by use of intuition and expert judgment, models can be used to test assumptions and examine the impact of policies and resource levels. The autonomous nature of responders as well as the rapidly changing information during a disaster suggests that agent based models can be especially suited for examining policy questions. In this work, we built an agent based model of a given urban area to simulate the emergency medical response to a mass casualty incident (MCI) in that area. The model was constructed from publicly available geographic information system and data regarding available response resources (such as ambulances, EMS personnel and hospital beds). Three different agent types are defined to model heterogeneous entities in the system. By simulating various response policies, the model can inform emergency responders on the requirements and response protocols for disaster response and build intuition and understanding in advance of facing actual incidents that are rare in the day-to-day operating experiences. Show less

Abu Dhabi Health Services (SEHA) has established Disease Prevention and Screening Centers (DPSCs) to provide health screening for third-country national (TCN) workers. In anticipation of increases in the number of TCN workers who will require health screening, SEHA is building new centers to accommodate the increased requirements. As part of the design process, we used queuing and simulation models to model (1) the current DPSC configuration, (2) a configuration based on individual lanes, and… Show more

Abu Dhabi Health Services (SEHA) has established Disease Prevention and Screening Centers (DPSCs) to provide health screening for third-country national (TCN) workers. In anticipation of increases in the number of TCN workers who will require health screening, SEHA is building new centers to accommodate the increased requirements. As part of the design process, we used queuing and simulation models to model (1) the current DPSC configuration, (2) a configuration based on individual lanes, and (3) a configuration based on pooled lanes. We then used these models to analyze the effects of design decisions on capacity and customer flow through the centers. Show less

The modeling and simulation of inpatient healthcare systems comprising of multiple interconnected units of monitored care is a challenging task given the nature of clinical practices and procedures that regulate patient flow. Therefore, any related study on the properties of patient flow should (i) explicitly consider the modeling of patient movement rules in face of congestion, and (ii) examine the sensitivity of simulation output, expressed by patient delays and diversions, over different… Show more

The modeling and simulation of inpatient healthcare systems comprising of multiple interconnected units of monitored care is a challenging task given the nature of clinical practices and procedures that regulate patient flow. Therefore, any related study on the properties of patient flow should (i) explicitly consider the modeling of patient movement rules in face of congestion, and (ii) examine the sensitivity of simulation output, expressed by patient delays and diversions, over different patient movement modeling approaches. In this work, we use a high fidelity simulation model of a tertiary facility that can incorporate complex patient movement rules to investigate the challenges inherent in its employment for resource allocation tasks. Show less

Health care capacity decisions are often based on average performance metrics such as utilization. However, such decisions can be misleading, as a large portion of the costs in service operations is due to the inability to provide service due to congestion. This paper will review sources of variation that affect in-patient care capacity and develop a series of models of patient flow in a health care facility. We demonstrate that even in settings where the patient population and services… Show more

Health care capacity decisions are often based on average performance metrics such as utilization. However, such decisions can be misleading, as a large portion of the costs in service operations is due to the inability to provide service due to congestion. This paper will review sources of variation that affect in-patient care capacity and develop a series of models of patient flow in a health care facility. We demonstrate that even in settings where the patient population and services provided are fixed, models that do not account for natural variations in the arrival rate and correlation in patient lengths of stay in sequential units will show the same utilization, but underestimate congestion and the resulting costs. Therefore, we argue that utilization is an inappropriate measure for validating models and congestion metrics such as blocking and diversions should be used instead. Show less

We propose an indifference-zone approach for a ranking and selection (R&S) problem with the goal of finding the best-subset from a finite number of competing simulated systems given a level of correct-selection probability. Here the “best”system refers to the system with the largest or smallest performance measures. We present a best-subset selection procedure that can effectively eliminate the non-competitive systems and return only those alternatives as the selection result where statistically… Show more

We propose an indifference-zone approach for a ranking and selection (R&S) problem with the goal of finding the best-subset from a finite number of competing simulated systems given a level of correct-selection probability. Here the “best”system refers to the system with the largest or smallest performance measures. We present a best-subset selection procedure that can effectively eliminate the non-competitive systems and return only those alternatives as the selection result where statistically confident conclusions hold. Numerical experiments document that our procedure works well by selecting the correct best-subset with very high probability. Show less

Effective scheduling of elective cases in an operating room suite is a challenging task due to inherent uncertainty and competing performance metrics. In this paper, we present a simulation model for the surgical suite within the VA Pittsburgh Health Care System (VAPHS) that is used to evaluate and optimize different scheduling policies. A flexible set of probabilistic scheduling rules is evaluated and a dynamic scheduling policy is proposed as an alternative to static strategies. The dynamic… Show more

Effective scheduling of elective cases in an operating room suite is a challenging task due to inherent uncertainty and competing performance metrics. In this paper, we present a simulation model for the surgical suite within the VA Pittsburgh Health Care System (VAPHS) that is used to evaluate and optimize different scheduling policies. A flexible set of probabilistic scheduling rules is evaluated and a dynamic scheduling policy is proposed as an alternative to static strategies. The dynamic scheduling policy allows us to reduce the variance in patient waiting times and backlogs. The developed simulation model is based
on the data collected at the VAPHS. Show less

Iravani, S.M.R., T. Liu, K. L. Luangkesorn, and D. Simchi-Levi, (2007), “Produce-to-Stock Systems with Advance Demand Information and Secondary Customers”, in Naval Research Logistics, Vol. 54, No. 3, pp. 331-345.

Iravani, S.M.R., K. L. Luangkesorn, and D. Simchi-Levi, (2004), “A general decomposition algorithm for parallel queues with correlated arrivals.” Queueing Systems Vol. 47, No. 4. pp. 313-344.

Iravani, S.M.R., K. L. Luangkesorn, and D. Simchi-Levi (2003), “On assemble-to-order systems with flexible customers”, IIE Transactions Vol. 35, No. 5. pp. 389-403.