This paper introduces a new generic problem to the literature of Workforce Scheduling and Routing Problem. In this problem, multiple workers are assigned to a shared vehicle based on their qualifications and customer demands, and then the route is formed so that a traveler may be dropped off and picked up later to minimize total flow time. We introduced a mixed-integer linear programming model for the problem. To solve the problem, an Adaptive Large Neighborhood Search (ALNS) algorithm was developed with problem-specific heuristics and a decomposition-based constructive upper bound algorithm (UBA). To analyze the impact of newly introduced policies, service area, difficulty of service, distribution of care, and number of demand nodes type instance characteristics are considered. The empirical analyses showed that the ALNS algorithm presents solutions with up to 35% less total flow time than the UBA. The implementation of the proposed drop-off and pick-up (DP) and vehicle sharing policies present up to 24% decrease in total flow time or provide savings on the total cost of service especially when the demand nodes are located in small areas like in urban areas.

翻译：本文在劳动力调度与路径规划问题领域提出了一种新型通用问题。在该问题中，多名工人根据其资质与客户需求被分配至共享车辆，进而规划路径，使得作业者可中途下车并在后续被接回，以最小化总流程时间。我们为该问题建立了混合整数线性规划模型。为求解该问题，我们开发了自适应大邻域搜索算法，该算法结合了问题特定启发式策略与基于分解的构造性上界算法。为分析新引入策略的影响，研究考虑了服务区域、服务难度、服务分布及需求节点数量等实例特征。实证分析表明，相较于上界算法，自适应大邻域搜索算法解的方案可实现总流程时间降低高达35%。所提出的上下客策略与车辆共享策略的实施可使总流程时间减少最多24%，或在需求节点集中于小范围区域（如城市区域）时有效节约总服务成本。