Dynamic stochastic vehicle routing model in home healthcare scheduling.

摘要

Home Healthcare is a growing sector in the medical service business and involves visiting and nursing patients at their homes. Senior Care (SC), a department within the MU Sinclair School of Nursing currently operates with aides who are scheduled to visit patients on a daily basis at their homes. The SC department employs 18 aides and serves approximately 450 patients in Boone County and other areas of Missouri. The schedules assigned to these aides are generated manually, lacking dynamic capability and an overall inefficiency in the distribution of workloads among aids because of the unknown service times encountered at the patients home.;The framework adopted by this research is to formulate the above problem into a Dynamic Stochastic Vehicle Routing Problem (DSVRP) and to solve using heuristics techniques. Cluster First Route Second heuristic is applied to solve the VRP and to generate a-priori solution. Three new classes of re-routing strategies are proposed to handle VRP in dynamic environment. The 3 strategies were each tested for 2 different operating scenarios and verified by performance evaluation, with efficiency and overtime minimization as the criterion. The research conducted in this thesis can be used as a test bed upon which future research in the area of practical dynamic systems can be tested and validated.