Design And Implementation Of Food Delivery Based On Multi-level Workers

With the rapid development of mobile internet,food delivery based on spatial crowdsourcing is becoming more and more popular.Spatial crowdsourcing platforms integrate orders from various restaurants and worker resources,and assign suitable tasks based on worker’s location and personal preferences.Existing works have conducted in-depth research on food delivery based on spatial crowdsourcing and proposed many efficient algorithms,but they have all ignored the community constraints of workers.To avoid unnecessary time consumption,the road network and workers are divided into many communities in practice,and tasks are divided into in-community tasks and cross-community tasks.Commnity constraints refer to the fact that workers can only be assigned tasks starting from restaurants located in their respective communities.Therefore,workers have to need to immediately return to their respective communities after completing cross-community tasks to participate in subsequent task assignment,which seriously reduces their efficiency.Based on this,this paper proposes a food delivery scheme based on multi-level workers,which introduces taxi workers with weak coixmmunity constrains that are already present in the road network,cooperating with existing food delivery workers to complete cross-community tasks in a relay manner.The main work of this paper is summarized as follows:Research on Task Assignment based on Multi-level Workers.This paper models the task assignment of multi-level workers as a hybrid dimension matching problem,In order to match as many tasks as possible and minimize the overall travel cost of workers,this paper proposes multi-level worker scheduling algorithms based on 2dimensional matching and 3-dimensional matching,and finally designs a network flow framework to achieve eficient hybrid dimension matching solutions.Experimental results show that compared with traditional single-level worker delivery plans,the multi-level worker scheduling scheme achieves better results in task assignment rate and average driving distance per worker on the platform.Research on Revenue Allocation based on Heterogeneous Platform Cooperation.This paper uses cooperative game modeling to calculate the Shapley value of each platform for efficient and reasonable revenue allocation between the food delivery platform and the taxi platform.In order to improve the computational efficiency of revenue allocation,this paper proposes a revenue calculation method based on incremental task allocation and an effective pruning strategy,respectively.The experimental results show that compared with the traditional average revenue allocation method,the Shapley value-based revenue allocation method satisfies the individual rationality of the cooperation platform and realizes a more fair and reasonable revenue allocation.To provide sufficient taxi workers,the food delivery platform can cooperate with multiple taxi platforms.The food delivery platform provides tasks,and each taxi platform provides workers.After the workers complete the tasks,the revenue is shared by the food delivery platform and the taxi platform.Fair and reasonable revenue allocation ensures the interests of each platform and motivates them to continue to participate in cooperation.However,existing revenue allocation methods cannot quantify the true contribution of each cooperative platform.This paper models it through cooperative game theory,calculates the Shapley value of each platform to allocate revenue,and proposes a revenue calculation method based on incremental task matching and effective pruning strategies to improve the efficiency of revenue allocation.Experimental results show that compared with traditional equal revenue allocation,the method based on Shapley value satisfies the individual rationality of cooperative platforms and achieves more fair and reasonable revenue allocation.Overall,this paper design and also implement a food delivery scheme based on multi-level workers,and conducts research on task allocation and revenue allocation.The scheme achieves efficient multi-level workers collaboration and fair heterogeneous platform revenue allocation,enhancing the efficiency of workers and creating greater revenue.