A location–allocation model for service providers with application to not-for-profit health care organizations
Introduction
At the beginning of the 21st century, health care remains one of the areas of crucial concern for millions of Americans. As evidence, it may be noted that obtaining and paying for high-quality health care was a major topic of debate in both the 2004 and 2008 presidential elections. While the United States is fortunate in possessing the world's finest medical facilities and technologies, it is well-known that health care costs are rapidly spiraling upward, currently constituting some 15% of national spending [1].
Health care providers are pressured by two conflicting aspects of the public debate on health issues. On one hand, it is clear that Americans value the quality of their health care facilities, treatments, and options. On the other hand, much to the consternation of the American public, the continuously rising cost of health care is reflected in ever-increasing health care insurance premiums, which remain a significant expense for most households. At a time when medical procedures, new technologies, and drugs are highly developed and consequently very expensive, the pressure to control costs is affecting the operations of health care organizations in general and hospitals in particular. While the public debate about health care will not diminish, health care providers such as hospitals have no choice at present but to become as efficient as possible in all aspects of their operations [2], [3].
This study is motivated by the need for effective delivery of specialized health services at the Department of Veterans Affairs (VA), the primary service organization dedicated to the care and well-being of American veterans. In particular, this study is based upon a funded research project that investigated the optimal location of traumatic brain injury (TBI) treatment units for VA medical centers. As a not-for-profit service organization, the VA has to define optimality using a multi-objective approach where both controlling the cost of providing care and determining the level and extent of the care provided (i.e., access and availability) are equally important goals to pursue.
This paper is organized as follows. Section 2 provides the motivation and literature review germane to our research. Section 3 describes our optimization model and defines the relevant decision variables and model coefficients. Section 4 describes our solution methodology for the optimization model while Section 5 provides the results of a rigorous computational testing of the model. Section 6 applies our model in a real-world experimental setting based upon our work with the VA. Last, concluding remarks and future research directions are given in Section 7.
Section snippets
Motivation and literature review
Location–allocation models seek to simultaneously determine optimal facility locations and the assignment of customers to the selected facilities. The research literature in facility location is vast. For the interested reader, a comprehensive review of general facility location models and the methods used to solve them may be found in Love et al. [4] and Cornuéjols et al. [5]. Location–allocation models have been extensively applied in health care settings including locating hospitals in rural
The optimization model
The purpose of the research presented here is to develop a framework for mathematically analyzing the issues related to the location of specialized health care services for the VA. The optimization model presented in this section is based on the needs of the VA as a not-for-profit service organization. It includes two primary criteria: (1) the VA's cost of providing service and (2) the service rate provided to the VA's patients. The cost of service is made up of fixed and variable treatment
Solution methodology
Model (VA) contains a large number of binary variables and also a large number of constraints. For example, if our model were to be applied on a national scale, the resulting application would contain approximately 43,000 districts at the ZIP code level and 155 possible treatment unit locations corresponding to the current total number of medical centers within the VA. The presence of the 0–1 variables and the potentially large model size puts the model into the category of combinatorial
Computational experiment
The computational experiment was carried out on a Compaq Presario desktop computer with 512 MB of memory running at 2.2 GHz on an AMD Athlon processor. As stated earlier, Model (VA) is NP-complete [14] and the solution times tend to increase exponentially when searching for the optimal solution with general-purpose solvers such as Lingo 7. In our investigations, we found that solving problems with 25 candidate medical center locations, 5 open treatment units, and 40 patient districts (the same
Managerial experiment and implications
Service organizations such as the VA are under great pressure along a number of conflicting fronts. While some of these are well documented, others are more subtle and related to the internal dynamics of a particular organization. In fact, some of the criteria might not even be explicitly acknowledged by an organization but nevertheless exert real influence on the decision-making process (in this case, with regard to location and capacity issues). Our work with some of these organizations, not
Conclusion
In this paper, we have developed a model for the location of specialized health care services such as TBI treatment units. We have applied this model to data based on one of the VA's integrated service networks. The model incorporates two primary criteria for the location of treatment units: (1) a cost minimization objective that includes relevant costs such as treatment, transportation, labor, and fixed costs, and (2) a minimum service proportion requirement that is consistent with the VA's
Acknowledgments
This work was funded by the Department of Veterans Affairs (VA) Rehabilitation Research and Development (RR&D) Service through the VA RR&D Rehabilitation Outcomes Research Center at the Malcolm Randall VA Medical Center in Gainesville, Florida (Project #B2610R, W. Bruce Vogel, Principal Investigator).
Disclaimer: The views expressed here are those of the researchers and do not necessarily reflect the views of the Department of Veterans Affairs.
References (24)
- et al.
Planning health services with explicit geographical considerations: a stochastic location–allocation approach
Omega
(2005) - et al.
Location and service mix decisions for a managed health care network
Socio-Economic Planning Sciences
(1993) A model and methodologies for the location problem with logistical components
Computers and Operations Research
(2002)- et al.
Efficient formation of storage classes for warehouse storage location assignment: a simulated annealing approach
Omega
(2008) - ...
- et al.
Review of operations research improvements in patient care delivery systems
(1989) - et al.
Capacity management in health care services: review and future research directions
Decision Sciences
(1988) - et al.
Facilities location: models and methods
(1988) - et al.
The uncapacitated facility location problem
On the implementation of quantitative facility location models: the case of a hospital in a rural region
Journal of the Operational Research Society
(1996)
Locating a blood bank
Interfaces
A trauma resource allocation model for ambulances and hospitals
Health Services Research
Cited by (101)
Optimal trauma care network redesign with government subsidy: A bilevel integer programming approach
2023, Omega (United Kingdom)Spatial multi-objective optimization of institutional elderly-care facilities: A case study in Shanghai
2023, International Journal of Applied Earth Observation and GeoinformationAn exact criterion space search method for a bi-objective nursing home location and allocation problem
2023, Mathematics and Computers in SimulationAn active-learning Pareto evolutionary algorithm for parcel locker network design considering accessibility of customers
2022, Computers and Operations ResearchTowards an Optimized and Sustainable Blood Supply Chain Network under Uncertainty: A Literature Review
2022, Cleaner Logistics and Supply Chain
- 1
Tel.: +1 303 724 2416; fax: +1 303 724 2530.