Robust Optimization for Home Health Care

One of the vital topics in modern societies is the improvement of life expectancy. Care providers are dealing with a number of challenging issues (labor-intensiveness, growing operation costs, increasing demand for satisfaction level of both patients and workforce associations, the regulations regarding working time) to provide best service for residents. In attempt to satisfy all the desires of patients, nurses, and care providers within a operation cost limit, it is, therefore, worth considering all the operational decisions affecting the efficiency of the health care services - with particular emphasis on home health care services in our case - within a single optimization framework. Having optimal solutions on service operation can play a potential role in offering patients high quality medical service as well as satisfying the demands of nurses and care providers. This topic is currently very popular because of its importance in improving the quality of life, and also because it leads to a considerable reduction in expenditure. Consequently, the problem has become an interesting topic, and caught the special attention of researchers and organizations during the last few decades. Most of the algorithmic contributions in solving the problems arising in health care services are exact methods, which are only suitable for small-sized problems (but providing optimal solutions), and heuristic algorithms which are able to handle larger instances (but there is no guarantee about the solution quality). In addition, most of the studies have been carried out under the context of classical optimization in which parameters are assumed to be known precisely in advance. However, in reality, the parameters are difficult to know precisely, because of the changing nature of the environment. When this uncertainty factor is ignored, the optimization would be done according to the nominal (i.e. estimated, or assumed) parameter values, and this would provide solutions which are optimal only according to the mathematical model at hand, but not according to the reality. As a result, such an "optimal" solution might turn out to be very costly, or even infeasible, when implemented in reality. This research will focus on developing effective optimization algorithms based on the integration or hybridization of matheuristics (a combination of exact algorithms and metaheuristics) and robust optimization (a promising methodology used to handle the uncertainty of the problem parameters) to solve effectively issues arising in home health care services under the context of uncertainty.