| dc.description.abstract | This paper studies the vehicle routing and scheduling problems with time constraints for
the regular health care system that provides healthcare at residence for the elderly. The
objective of this study is to optimize workforce assignment and satisfy patients’ time
requirements by minimizing the operation time and workforce usage. The workforces
are mobile clinic vehicles that visit the patients and are based at one depot. The vehicles
start daily trips at a hospital depot, travel to locations with demand to perform
examinations, and return to the depot within a working day. The model has four
processes: route generation, route scoring and selection, planning generation, and
planning improvement. The model uses a simulated annealing process to generate route
sets, a score-based system to select routes, a distance-based system to decide the
planning order, and a genetic algorithm with uniform crossover and one-point crossover
is used to find improvement in the total operation time. Benchmark and generated data
are used as inputs. The shortest total operation time recorded is 26,955.5 minutes and
16,962.82 minutes for benchmark and generated data, respectively. The average
improvement the genetic algorithm has on the benchmark result is 1.097% but none for
the generated data. Extensions on traveling limits and different order planning methods
are discussed in the generated data section. The normal order and descending distance
methods perform similarly. The travelling limit generates the best at 100 minutes per
route, which improves the result by 0.85%. Further research is needed to evaluate the
economical value of the model. | en_US |
