The Roadside Healthcare Facility Location Problem



The population of truck drivers plays a key role in the spread of HIV and other infectious diseases in sub-Saharan Africa. Truck drivers thereby affect the health and lives of many, but also suffer from poor health and significantly reduced life expectancy themselves. Due to professional circumstances, their health service needs are generally not well addressed. Therefore, NGO North Star Alliance builds a network of healthcare facilities along the largest trucking routes in sub-Saharan Africa. Choosing the locations of new facilities presents novel and complex optimization problems. In this talk, we consider a general network design problem: the Roadside Healthcare Facility Location Problem (RHFLP). The RFHLP entails to select locations for facilities and to assign (additional) service packages to them. The objective combines the maximization of the patient volume at these facilities and the maximization of the effectiveness of the health service delivery to the population served. The latter criterion is modeled through three novel access measures which capture the needs for effective service provisioning. The resulting optimization problem is essentially different from previously studied healthcare facility location problems, because of the specific mobile nature of health service demand.  Applying our model to the network of major transport corridors in South-East Africa, we investigate several prominent questions managers and decision makers face. We show that the current practice of maximizing patient volumes may need to be reconsidered, as substantial gains in effectiveness can be made when allowing a small reduction in patient volumes. We furthermore show that solutions are rather robust to data impreciseness and that long term network planning can pay off substantially, particularly in greenfield situations.