An Optimization-Based Prototype for Nurse Assignment Prattana Punnakitikashem† and Jay M. Rosenberger Department of Industrial and Manufacturing Systems Engineering, The University of Texas at Arlington, Arlington, TX, 76019, USA +1817-272-3092, Email: {prattana.punnakitikashem, jrosenbe}@uta.edu Deborah Buckley Behan Robert L. Baker Kimberly Goss School of Nursing, The University of Texas at Arlington, Arlington, TX, 76019, USA +1817-272-4860, Email:
[email protected],
[email protected],
[email protected]
Abstract. One of the major problems in the United States health care system is a shortage of nurses. We describe four stages of nurse planning. Our research focuses on the last stage that makes daily decisions on assigning nurses to patients. We develop an information technology (IT) prototype for assigning nurses to patients with the purpose of minimizing excess workload on the nurses. We present the structure of the IT prototype and summarize an underlying optimization model. Finally, we discuss areas of future research. Keywords: Nurse assignment, nurse planning, nurse assignment IT prototype.
1. INTRODUCTION One of the major problems in the United States health care system today is a shortage of nurses. There were 118,000 vacant positions for Registered Nurses (RNs) in April 2006 (AACN 2002). Baby boomers (AACN 2002) and an increasing elderly population, who requires a substantial amount of health care, are the main reasons for a rapid growing demand of nurses (Victor and Higginson 1994). Despite the current situation, the enrollments in RN degree programs declined by 50,000 nurses from 1993 to 2001 (Lynn 2001). With fewer new nurses entering the profession, the average age of working RNs is increasing (Buerhaus et al. 2000). In the United States, more than 40% of nurses have suffered from work burnout and one-fifth of all nurses planned to leave their job within next year (Aiken et al 2001). Due to the current low retention and low entering rates of nurses, the health care industry will need more than 1.2 million new and replacement nurses by 2014 (Hecker 2005). Buerhaus et al. (2000) predicted that by 2020 the United States will face a 20% shortage in the number of nurses needed in the nation’s health care system. The shortage of nurses has a direct effect on patient care. The most important factors, which contribute to medical error, are workload, stress, and fatigue of health
________________________________________ † : Corresponding Author
professionals; not enough time spent with patients; and not enough nurses in the health care system (KFF 2004). Powers (1993) observed that excessive workload enhances poor quality of patient care. Patients with insufficient care have higher failure-to-rescue rates and risk adjusted 30-day mortality when they are cared for by nurses with too many assigned patients (Aiken et al. 2002). The nursing shortage not only affects patients, but it also has a significant impact on the quality of nursing work. Having received too many patients, nurses are more likely to suffer from job dissatisfaction and burnout (Aiken et al 2002). One way to ease the burden of nursing shortage and improve the quality of nursing work is to balance the workload of nurses or reduce the excessive workload on nurses. A nurse-patient assignment is a mandatory routine for all healthcare units in almost every hospital in the world, and it is performed daily for every shift for the entire year. At the beginning of a nursing shift, a charge nurse assigns each nurse to a set of patients for a shift. Since some patients require more care than others, tending to the needs of a few patients may consume most of a nurse's time, while other patients may receive only minimal care. During the same shift another nurse may have significantly less workload because his (her) patients may require less care. The second nurse should assist the first by taking on some
of the first nurse's patients. Similarly, there can also be differences in the skills of the nurses, so assigning the right nurses to the right patients can reduce excess workload. Currently a nurse-patient assignment is done manually; therefore a computerized assignment tool can help a charge nurse with a cumbersome time-consuming task. In this research, we developed an optimization-based prototype to propose an assignment of nurses to patients in a hospital unit. The purpose of the IT prototype is to minimize excess workload on the nurses, which enhance quality of patient care, increase the quality of nursing work, and ultimately reduce the burden of nursing shortage. In Section 2, we describe the four stages of nurse planning and their related literatures. In Section 3, we summarize the underlying nurse assignment model of the IT prototype. In Section 4, we present the structure of the IT prototype. In Section 5, we describe an implementation and training of our IT prototype to potential users. In Section 6, we summarize the results of the implementation and training. Finally, we discuss the conclusions and areas of future research of the IT prototype for nurse assignment in Section 7.
2. NURSE PLANNING LITERATURE
AND
RELATED
There are four stage of nurse planning: nurse budgeting, nurse scheduling, nurse rescheduling, and nurse assignment. In the first stage, nurse budgeting, the number of nurses needed during a year is forecasted to create a budget. Trivedi (1981) developed the mixed integer goal programming model. Kao and Queyranne (1985) showed that a singleperiod demand estimate provided good approximation to the nursing budget for a hospital nursing unit. Martel and Ouellet (1994) applied the stochastic programming to allocate the budget of a nursing unit to different types of nurses. The second stage of nurse planning is nurse scheduling. A nursing administrator forecasts the number of patients that will enter the hospital. Based upon the predicted number of patients, the number of nurses can be determined by various approaches. Burke et al. (2004) stated that nurse scheduling can be classified by a time horizon in which decisions are made. We refer to nurse staffing as a long term workforce management decision made while nurse rostering describes the short term (several weeks) allocation of nurses to a working time period. Most nurse rostering literature can be found in Cheang et al (2003) and Burke et al. (2004), providing nurse rostering survey papers. They summarized the overview of the nurse rostering model and the solution methodologies from the 1960’s until 2004. Most academic literature on nurse planning is on rostering (Warner and Prawda 1972, Warner 1976, Tien and Kamiyama 1982, Okada and Okada 1988, Bradley and Martin 1990, Chen and Yeung
1993, Hung 1995, Jaumard et al. 1998, Aickelin and Dowsland 2000, Burke et al. 2002, and Bellanti et al. 2004). These models only consider the nurse budgeting and scheduling stages, they ignore changes in staff and patient forecasts, and they assume the schedule will be followed as planned. Anecdotal evidence suggests that changes to the schedule are frequent, so intelligent decision support to reschedule nurses will likely improve nurse planning. The third stage of nurse planning, nurse rescheduling, involves revising the set of nurses scheduled for a shift. Approximately 90 minutes before each shift, a nurse supervisor reevaluates the scheduled nurses based upon the activities of the previous shift, the patients in the emergency room, and either a census matrix or a patient classification system. If more nurses are needed than are scheduled for the upcoming shift, the supervisor attempts to recruit additional nurses who work as needed--PRN nurses, nurses who work part time--part-time nurses, and nurses who are not scheduled for the upcoming shift--off duty nurses. If an insufficient set of nurses agree to work the shift, the supervisor, upon approval from the nurse manager, hires temporary agency nurses to satisfy the remaining shortage. If more nurses are scheduled for the shift than needed, then the supervisor instructs surplus PRN nurses and part-time nurses to take the day off without pay. Patient classification systems are the most sophisticated technology for nurse rescheduling. These systems group patients into one of several categories. They estimate how many times certain tasks will be performed in caring for a patient in each category. Using these estimates and the expected time required to perform each task, the systems determine the amount of time to care for a typical patient. As a patient's condition changes, (s)he may be given a new patient classification. Siferd and Benton (1994) determined the number of nurses needed for a shift based upon the number of patients, mean patient acuity, and rate of patient acuity change. Nevertheless, the different sets of skills among the nurses are ignored. Bard and Purnomo (2005) developed an integer programming model for daily nurse rescheduling with nurse preference. Vericourt and Jennings (2006) employed a queuing model to investigate nurse-to-patient ratios mandated by California State and they proposed two heuristic staffing policies. In the fourth stage, nurse assignment, the charge nurse assigns patients to nurses at the beginning of a shift. Modern patient classification systems partition the set of patients into groups, and each group is assigned to a nurse (Overfelt 2004). Walts and Kapadia (1996) presented a patient classification system, which was based on a linear programming model, to determine the minimal number of nurses to meet the required workload level, but they did not use a detailed nurse assignment model. Mullinax and Lawley (2002) developed a neonatal acuity system to determine a nurse workload per
patient and then used an integer linear programming model to propose patients-nurse assignment in a neonatal intensive care unit. Punnakitikashem et al. (2005) developed a twostage stochastic integer programming model for a nursepatient assignment that considered uncertainty in patient care. Our model objective was to minimize excess workload for nurses. Sundaramoorthi et al. (2006) presented a simulation model from real data to evaluate nurse-patient assignments. We are unaware of any computerized tools that are used to perform a nurse-patient assignment.
3. NURSE ASSIGNMENT MODEL In this section, we describe an underlying model that assigns nurses to a set of patients for a nurse shift. We classify patient care into two types: direct care and indirect care. We refer to direct care as the amount of time nurses spend directly with patients, for example, giving medicine to patients, measuring blood pressure for patients, or bathing patients. Indirect care is defined as the amount of time nurses spend on other tasks for patients, such as documentation of patients’ condition and coordinating their care. The total workload is the total amount of time a nurse spends to care for all of her patients in a particular time period, which is comprised of both direct care and indirect care. Excess workload is the difference between the total workload and the amount of time a nurse is able to work in a time period. In reality, when a nurse is assigned excess workload, then (s)he must find assistance from other nurses or postpone some of her workload. The objective of our model is to minimize the excess workload on nurses. We model the nurse assignment as a mixed-integer programming problem. The model uses decision variables, constraints, and input parameters based upon the nurses, the patients, and the shift information. More details about the underlying model can be found in Punnakitikashem et al. (2005) as the mean value problem; nevertheless we briefly introduce the model as follows. The shift is partitioned into a set of time periods, and the input parameters include the amount of direct and indirect care that a nurse must provide to a patient in a time period if an assignment occurs. We obtained a direct care and indirect care parameter from mined encrypted data from Baylor Regional Medical Center, Grapevine, TX, USA. More details about input parameter also can be found in Punnakitikashem et al. (2005). For each patient-nurse pair, we create an assignment decision variable, which is a binary variable. The value of this variable for a specific patient-nurse pair is equal to one if that patient is assigned to that nurse, or zero otherwise. For each time period and each nurse, we have a workload decision variable to represent the total amount of patient care the nurse performs in the time period. We assume that direct care must be performed in the given time period, but indirect
care can be performed through out the shift. Consequently, we create indirect care decision variables that determine when indirect care is performed. In the mathematical description of the model, there are three constraints. The first constraint set is an assignment constraint ensuring that every patient must be assigned to one of the nurses. The second constraint set are the total workload constraints; the workload of a nurse during a given time period included both direct care and indirect care. Finally, an indirect care constraint allows indirect care to be performed from the beginning of a given time period until the end of the shift. As a nurse’s workload increases, her patients receive less care, so the model minimizes a function that penalizes the assignment of excess workload on a nurse. The penalty for assigning more work to an overworked nurse is greater than assigning that to a nurse with a lighter workload. Minimizing this function naturally balances the workload. The function penalizes an assignment in which a nurse must work more than (s)he is able to work during any time period. The objective of our model is to minimize the excess workload on nurses.
4. IT PROTOTYPE STRUCTURE In this section, the structure of IT prototype is presented. The IT prototype decision support system involved two software programs, Microsoft Excel and WinSCP, in a charge nurse personal computer. WinSCP is a freeware SFTP (Secure Shell file transfer protocol) for Microsoft Windows (WinSCP 2006). The IT prototype includes three main functions for the users: 1. Shift information entry The interface of the IT prototype is in Microsoft Excel. At the beginning of the shift, a user enters shift information containing nurses’ and patients’ information to a Microsoft Excel spreadsheet as follows: Shift information: • Shift: day, evening, night, Patients’ information • Number of patients, • Patient rooms, • Primary diagnosis, • Specific nurse requirement, i.e., a patient requires a special care from a certain type of nurse, • Admission time (if known), • Discharge time (if known), Nurses’ information • Number of nurses, • Nurse type: registered nurse (RN), licensed vocational nurse (LVN), nurse aid (NA), chemotherapy nurse, and pediatric nurse.
Figure 1 shows the Microsoft Excel interface, which is a spreadsheet for shift information entry. The IT prototype uses the visual basic macros to create a batch file containing current shift information.
Figure 1: The shift information entry spreadsheet in Microsoft Excel 2. Data transferring WinSCP is used to transfer a batch file from a user personal computer to a Dual 3.06-GHz Intel Xeon Workstation. The underlying model is solved by using CPLEX 9.1 callable library. The workstation calls an executable file and returns an optimal assignment file. WinSCP is again used to transfer an assignment file to a user personal computer. 3. Optimal assignment display The optimal assignment output can be displayed in the Microsoft Excel spreadsheet with the visual basic macros. The IT prototype proposes a nurse-patient assignment specifying an assignment for each nurse to a group of her patients with the minimal excess workload for all nurses. Figure 2 displays the optimal assignment output spreadsheet for users. The architecture of the IT prototype is presented in Figure 3.
Figure 2: The optimal assignment spreadsheet in the Microsoft Excel.
User Interface – Microsoft Excel Input • Shift type • Number of patients • Number of nurses • Patient information (room, diagnosis, nurse requirement, admit time, discharge time) • Nurse information (type of nurses)
Output • Nurse-patient assignment
Shift information WinSCP
WinSCP
5. IMPLEMENTATION AND TRAINING UNIX workstation
In this section, we describe a training seminar for the IT prototype decision support system to potential users. Subjects for this seminar were nursing students at The
Figure 3: The IT prototype structure.
Assignment
University of Texas at Arlington (UTA). The IT prototype was installed at the computer lab in the School of Nursing at UTA. We provided a training course to use the system and collect feedback from them. The training course was given to two nursing research classes at UTA, which are required for RN-BSN students (undergraduate-level class) and MS students (graduate-level class) in the summer 2006 semester. The training curriculum included • • • • • •
A discussion of the factors involved in assigning nurses to patients, such as patient conditions and room locations, Background and motivation of the project, Descriptions of the collected data and mining results from the four units at Baylor, Overviews of the optimization models within the IT prototype, Demonstrations of the IT prototype on several examples, Feedback from the nursing students on the models, IT prototype, and training curriculum.
There were 20 and 13 subjects in the RN-BSN and the MS class, respectively. In the training classes, every subject received the following materials: nurse assignment presurvey, nurse assignment IT-prototype post-survey, presentation materials, a prototype instruction document, trouble shooting documentation, a census matrix, and a scenario for assignment. We began the training course with the pre-survey queried about their background on nursepatient assignment, computer skills, number of patients assigned for each shift, potential ways to improve a nurse assignment, etc. After the pre-survey, we gave a presentation of the above curriculum and followed by the IT prototype demonstration with an instruction documentation provided. Then, we simulated a nurse assignment in a medicalsurgery unit by providing subjects with a scenario of one shift in a medical-surgery unit and asked them to use the IT prototype program. The scenario contained patients’ information that described patients’ conditions, room location, admission time, discharge time, and type of nurse required. Subjects were asked to choose a number and a type of nurses from a census matrix. After having obtained shift information, subjects entered data and used the IT prototype to make an assignment. There were three research assistants to assist subjects if any problems occurred. One research assistant was in charge of technical support for the IT prototype while the other two nursing research assistants were responsible for nursing related inquiries. There was no time limit for subjects to use the systems. However, all subjects finished the assignment within 30 minutes. Subjects made a print out and submitted
their assignment. Post-surveys were given to subjects to evaluate the IT prototype based on level of difficulty, usefulness, features, and advantage-disadvantage. Subjects were also asked to list recommendations to improve the IT prototype. We collected all feedback from subjects. 6. SUMMARY OF RESULTS AND DISCUSSIONS In this section, we present an overview for results from the pre- and post-surveys of the IT prototype. There are two similar sets of surveys given to nursing students; one for RN-BSN and another one for MS students. We obtained feedback from various professions in the health care industry. Almost all of the MS students were employed in several health care units. Seventy percent of RN-BSN students and more than half of MS students experienced greater than or equal to four patients for a shift. Having high number of patients assigned, a nurse was likely to have job dissatisfactory and burnout. One-forth of RN-BSN and more than fifteen percent of MS students were not satisfied with the current assignment at work. We excluded one RN-BSN subject in the post-survey result because (s)he did not participate in the post-survey. More than forty percent of subjects reported that the IT prototypes were user friendly while thirty percent did not agree and the remaining did not answered. We obtained positive feedback from 15 out of 17 RN-BSN students (88.26%) and 11 out of 13 MS students (84.62%) supporting the use of IT prototype at their workplace. Two RN-BSN students were not eligible for this question because one worked in clinical setting and another one’s hospice made an assignment by demographics. The reason that one subject did not support the IT prototype at her work was that it was too time consuming. Another one subject did not answer this question. Subjects stated that the IT prototype benefited users for several reasons such as unbiased assignments, better assignments, speed, and decreased hand-written data. More details about survey results can be found at Baker et al. (2006). 7. CONCLUSIONS AND FUTURE RESEARCH In this research, we presented optimization-based prototype for assigning nurses to patients for a nursing shift with the minimal excess workload for all nurses. Instead of manually determining a nurse-patient assignment, it can now be promptly computerized by a charge nurse’s personal computer. We implemented the IT prototype and provided training courses to two groups of nursing students, RN-BSN and MS students at The University of Texas at Arlington. Subjects were asked to complete the pre- and post-survey to identify their background and
opinion about the IT prototype. More than four-fifths of the subjects had positive feedback supporting the use of the IT nurse-patient assignment prototype at their workplace. One interesting topic of future research is to incorporate acuity and continuity of care into the model. Lastly, the IT prototype with a nicer user interface and fewer steps could potentially attract more attention from nurses.
ACKNOWLEDGMENT This research is sponsored by Robert Wood Johnson Foundation under Grant Number 053963.
REFERENCES American Association of College of Nursing (2002) Nursing Shortage Fact Sheet. http://www.aacn.nche.edu/Media/FactSheets/NursingShortag e.htm Aickelin, U., and Dowsland, K. (2000) Exploiting problem structure in a genetic algorithm approach to a nurse rostering problem. Journal of Scheduling, 3(3), 139-153. Aiken, L.H., Clarke, S.P., Sloane, D.M., and Sochalski, J.A. (2001) Nurses’ reports of hospital quality of care and working conditions in five countries. Health Affairs, 20, 4353. Aiken, L.H., Clark, S.P., Sloane, D.M., Sochalski, J., and Silber, J.H. (2002) Hospital nurse staffing and patient mortality, nurse burnout, and Job Dissatisfaction. Journal of the American Medical Association, 288(16), 1987-1993. Bard, J.F., and Purnomo, H.W. (2005) Hospital-wide reactive scheduling of nurses with preference considerations. IIE Transactions, 35, 589–608. Baker, R.L., Behan, D.B, Rosenberger, J.M., Goss K., and Punnakitikashem, P. (2006) A pilot study to evaluate a computerized prototype used to generate hospital staff nurses a patient/shift assignment. Working paper. Bellanti, F., Carello, G., Croce, F.D., and Tadei Bellanti, F., Carello, G., Croce, F.D., and Tadei, R. (2004) A greedybased neighborhood search approach to a nurse rostering problem. European Journal of Operational Research, 153(1), 28-40. Bradley, D., and Martin, J. (1990) Continuous personnel scheduling algorithms: a literature review. Journal of the Society of Health Systems, 2, 8-23. Buerhaus, P.I., Staiger, D.O., and Auerbach, D.I. (2000) Implications of an aging registered nurse workforce. The Journal of the American Medical Association, 283(22), 2948-2954. Burke, E.K., Causmaecker, P.D., Petrovic, S., and Berghe, G.V. (2002) A multi criteria meta-heuristic approach to nurse rostering. Proceedings of Congress on Evolutionary Computation, Honolulu, IEEE Press, 1197-1202.
Burke, E. K., Causmaecker, P. D., Berghe, G. V., and Landeghem, H. V. (2004) The state of the art of nurse rostering. Journal of Scheduling, 7, 441–499. Centeno, M.A, Giachetti R., Linn R., and Ismail A. M. (2003) A simulation-ILP based tool for scheduling ER staff. Proceedings of the 2003 Simulation Conference: Driving Innovation, New Orleans, LA, 2, 1930–1938. Cheang, B., Li, H., Lim, A., and Rodrigues, B. (2003) Nurse rostering problems-a bibliographic survey. European Journal of Operational Research, 151, 447–460. Chen, J.G., and Yeung, T. (1993) Hybrid expert system approach to nurse scheduling. Computers in Nursing, 183192. Evans G. W., Gor T. B., and Unger E. (1996) A simulation model for evaluating personnel schedules in a hospital emergency department. Proceedings of the Winter Simulation Conference, Coronado, CA, 1205–1209. Farrington P.A., Nembhard H. B., Sturrock D. T., and Evans G. W. (1999) Emergency Department Simulation and Determination of Optimal Attending Physician Staffing Schedules, Proceedings of the Winter Simulation Conference, Phoenix, AZ, 1532-1540. Hecker, D.E. (2005) Occupational employment projections to 2014. Monthly Labor Review, U.S. Bureau of Labor Statistics, 70-101. http://www.bls.gov/opub/mlr/2005/11/art5full.pdf Hung, R. (1995) Hospital nurse scheduling. Journal of Nursing Administration, 25(7/8), 21-23. Jaumard, B., Semet, F., and Vovor, T. (1998) A generalized linear programming model for nurse scheduling, European Journal of Operational Research, 107, 1-18. Kao, E.P.C., and Queyranne, M. (1985) Budgeting costs of nursing in a hospital. Management Science, 31, 608–621. Klafehn K. A., Owens D. L., Felter R. A., Vonneman N., and McKinnon C. J. (1989) Evaluating the linkage between emergency medical services and the provision of scarce resources through simulation. Proceeding of the Annual Symposium on Computer Applications in Medical Care, Washington, DC, 335–339. Kumar A. P. and Kapur R. (1989) Discrete simulation application - scheduling staff for the emergency room. Proceedings of the Winter Simulation Conference, Washington, DC, 1112–1120. Lynn, G. (2001) The nursing shortage: causes, impact and Innovative Remedies. http://edworkforce.hourse.gov/hearings/107th/fc/nurses92501 /lynn.htm. Martel, A., and Ouellet, J. (1994) Stochastic allocation of a resource among partially interchangeable activities. European Journal of Operational Research, 74(3), 528539. McHugh M. L. (1989) Computer simulation as a method for selecting nurse staffing levels in hospitals. Proceedings of
the Winter Simulation Conference, Washington, DC, 1121– 1129. Mullinax, C., and Lawley, M. (2002) Assigning patients to nurses in neonatal intensive care. Journal of the Operational Research Society, 53, 25–35. Okada, M., and Okada, M. (1988) Prolog-based system for nursing staff scheduling implemented on a personal computer. Computers and Biomedical Research, 21, 53-63. Overfelt, F. (2004) Nursing ratios–are they good or bad? Proceedings of the 16th Annual Society for Health Systems Management Engineering Forum, Orlando, Florida. Powers, J. (1993) Accepting and refusing assignments. Nursing Management, 24(9), 64-73. Punnakitikashem, P., Rosenberger, J.M., and Behan, D.B. (2005) Stochastic programming for nurse assignment. Technical Report COSMOS 05-01, The University of Texas at Arlington. http://ieweb.uta.edu/TechReports/COSMOS-05-01.pdf Siferd, S.P., and Benton, W.C. (1994) A decision model for shift scheduling of nurses. European Journal of Operational Research, 74, 519–527. Sundaramoorthi, D., Chen, V.C.P., Rosenberger, J.M., Kim, S.B., and Behan, D.B. (2006) A data-integrated simulation model to evaluate nurse-patient assignments. Technical Report IMSE 06-05, The University of Texas at Arlington. http://ieweb.uta.edu/TechReports/IMSE-06-05.pdf Tien, J., and Kamiyama, A. (1982) On manpower scheduling algorithms. Society for Industrial and Applied Mathematics, 24(3), 275-287. Trivedi, V.M. (1981) Mixed-integer goal programming model for nursing service budgeting. Operations Research, 29, 1019–1034. Vericourt, F.D., and Jennings, O.B. (2006) Nurse-to-patient ratios in hospital staffing: a queuing perspective. http://faculty.fuqua.duke.edu/%7Efdv1/bio/ratios3.pdf Victor, C.R., and Higginson, I. (1994) Effectiveness of care for older people: a review. Quality Health Care, 3, 210-216. The Henry J. Kaiser Family Foundation (KFF). (2004) National survey on consumers' experiences with patient safety and quality information. Publication No. 7209. http://www.kff.org/kaiserpolls/7209.cfm Walts, L., and Kapadia, A. (1996) Patient classification system: an optimization approach. Health Care Management Review, 21, 75-82. Warner, D.M., and Prawda, J. (1972) A mathematical programming model for scheduling nursing personnel in a hospital. Management Science, 19(4), 411-422. Warner, D.M. (1976) Scheduling nursing personnel according to nursing preference: a mathematical programming approach. Operations Research, 24, 842-856. WinSCP (2006) Freeware SFTP and SCP client for Windows http://www.winscp.com/
AUTHOR BIOGRAPHIES Prattana Punnakitikashem is a Ph.D. student in the Department of Industrial & Manufacturing Systems Engineering at The University of Texas at Arlington. She received a bachelor’s degree in Chemical Engineering from Chulalongkorn University, Thailand, and a Master’s degree in Industrial and Manufacturing Systems Engineering at the University of Texas at Arlington. She works as a graduate research associate at the Center on Stochastic Modeling, Optimization and Statistic (COSMOS) at UTA. Her email address and web address are and , respectively. Jay M. Rosenberger is an Assistant Professor in the Department of Industrial & Manufacturing Systems Engineering at The University of Texas at Arlington. He is a faculty member in the center on stochastic modeling optimization and statistics (COSMOS), and a co-founder and director of Healthcare Enterprise Research Center (HERC) at The University of Texas at Arlington. He received a Ph. D. from the School of Industrial and Systems Engineering at the Georgia Institute of Technology, where he was awarded first place Pritsker Doctoral Dissertation Award by Institute of Industrial Engineers. Previously he worked as an operations research and decision support consultant at American Airlines. His teaching and research interests include applications of mathematical programming, stochastic optimization, and simulation. His email and web address are and . Deborah Buckley Behan is a Clinical Instructor in the School of Nursing, at The University of Texas at Arlington. She teaches critical care in the generic BSN program. She received her bachelor’s degree from Southwest Missouri State University, her Master’s degree at the University of Oklahoma, and is currently a Ph.D. candidate in the Nursing program at Texas Woman’s University. Previously, she worked as a Registered Nurse at Baylor Regional Medical Center, Grapevine, Texas. Her primary work experience has been with open-heart and cardiothoracic surgical patients. Her email and web address are and . Robert L. Baker is a Master’s student in the Nursing Administration Program in the School of Nursing at The University of Texas at Arlington, where he also received a bachelor’s degree in nursing. He currently works in home health for Presbyterian Hospital of Dallas, and has worked in critical care as an ICU nurse at the Dallas Veteran’s Affair Medical Center. His email address is:
. Kimberly Goss is a dialysis nurse at Veterans Affairs Medical Center and currently pursing a MSN at The University of Texas at Arlington. She attended Southern University in New Orleans, Louisiana and graduated with a BS in Biology/Chemistry. Ms. Goss attended El Centro Junior College and received an ADN, and a BSN at The University of Texas at Arlington. Her email address is: .
