Nowadays, by growing aged population and appearance of chronic diseases, there is an increasing demand for healthcare services so countries spend a considerable amount of their budget in healthcare area. At the same time, due to resource limitations, planning for optimal utilization of resources seems necessary. For instance, lack of hospitals and human resources is a limiting factor in healthcare system. Therefore, to reduce utilization of clinic and hospital beds, one can provide service to patients in their home. As care workers may not be in healthcare centers, it is possible to assign care workers to routes. Patients usually need to different services, and each care worker has different skills. Consequently, care workers' skills play vital role in their assignment. In addition, home care organizations usually have multiple centers because patients' homes may be located in different areas. In the thesis, a new mathematical model has been provided for multi-depot routing and selecting proper combinations of care workers in home health care problem. The objective function of the proposed model aims to minimize traveling time and time window violation penalties. As the problem is a type of NP-Hard problems, a parallel algorithm which is based on simulated annealing and tabu search has been proposed for large-sized problems. To evaluate the performance of the proposed algorithm, several small-sized problems has been solved by the algorithm, and the results have been compared with exact methods. Additionally, the performance of the algorithm has been compared with simulated annealing and tabu search for large-sized problems. Results indicate that the proposed algorithm outperform the two other algorithm. For validation of the algorithm, benchmark problems have been solved by the proposed algorithm. Finally, in order to assess the applicability of the model a case study has been considered. It is expected that this study will assist managers of health care providing organizations in their decision making.