RESUMO
The unbalanced allocation of healthcare resources is a major challenge that hinders access to healthcare. Taking Shenzhen as an example, this study aimed to enhance equity in obtaining healthcare services, through measuring and visualizing the spatial accessibility of community healthcare centers (CHC), and optimizing CHC geospatial allocation. We used the number of health technicians per 10,000 to represent the CHC's service capacity, combined with resident points and census data to calculate the population the CHC needs to carry, and then analyzed the accessibility based on the Gaussian two-step floating catchment area method. In 2020, five regions in Shenzhen had better spatial accessibility scores: Nanshan (0.250), Luohu (0.246), Futian (0.244), Dapeng (0.226), and Yantian (0.196). The spatial accessibility of CHCs shows a gradual decrease from the city center to the edge, which is affected by economic and topographic factors. With the support of the maximal covering location problem model, we selected up to 567 candidate locations for the new CHC, which could improve Shenzhen's accessibility score from 0.189 to 0.361 and increase the coverage population by 63.46% within a 15-min impedance. By introducing spatial techniques and maps, this study provides (a) new evidence for promoting equitable access to primary healthcare services in Shenzhen and (b) a foundation for improving the accessibility of public service facilities in other areas.
RESUMO
In this paper, an airport ground service task assignment problem is studied. A task represents a service, which must be performed by one or multiple ground crew of a shift with required qualification/proficiency within a prescribed time period. For every assigned task, define "task priority" times "task duration" as the "benefit" generated. The objective is to maximize the summation of "benefit" for all the assigned tasks. The problem is modeled as an integer linear programming problem with mathematical formulation. A branch-and-price algorithm is proposed for solving the problem instances to optimality. To expedite the column generation process, an acceleration strategy is proposed. The computational results show that our proposed branch-and-price algorithm is capable of solving large-sized instances and the acceleration strategy is quite effective in reducing the computational time. Moreover, the impact of changing various characteristics of tasks and shifts on the performance of the algorithm is studied in detail with supporting computational experiments. In particular, the impact of reducing the qualifications is significant with 20.82% improvement in the objective value.
